forked from torvalds/linux
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathdm-mpath.c
2256 lines (1877 loc) · 55.5 KB
/
dm-mpath.c
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
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
/*
* Copyright (C) 2003 Sistina Software Limited.
* Copyright (C) 2004-2005 Red Hat, Inc. All rights reserved.
*
* This file is released under the GPL.
*/
#include <linux/device-mapper.h>
#include "dm-rq.h"
#include "dm-bio-record.h"
#include "dm-path-selector.h"
#include "dm-uevent.h"
#include <linux/blkdev.h>
#include <linux/ctype.h>
#include <linux/init.h>
#include <linux/mempool.h>
#include <linux/module.h>
#include <linux/pagemap.h>
#include <linux/slab.h>
#include <linux/time.h>
#include <linux/timer.h>
#include <linux/workqueue.h>
#include <linux/delay.h>
#include <scsi/scsi_dh.h>
#include <linux/atomic.h>
#include <linux/blk-mq.h>
#define DM_MSG_PREFIX "multipath"
#define DM_PG_INIT_DELAY_MSECS 2000
#define DM_PG_INIT_DELAY_DEFAULT ((unsigned) -1)
#define QUEUE_IF_NO_PATH_TIMEOUT_DEFAULT 0
static unsigned long queue_if_no_path_timeout_secs = QUEUE_IF_NO_PATH_TIMEOUT_DEFAULT;
/* Path properties */
struct pgpath {
struct list_head list;
struct priority_group *pg; /* Owning PG */
unsigned fail_count; /* Cumulative failure count */
struct dm_path path;
struct delayed_work activate_path;
bool is_active:1; /* Path status */
};
#define path_to_pgpath(__pgp) container_of((__pgp), struct pgpath, path)
/*
* Paths are grouped into Priority Groups and numbered from 1 upwards.
* Each has a path selector which controls which path gets used.
*/
struct priority_group {
struct list_head list;
struct multipath *m; /* Owning multipath instance */
struct path_selector ps;
unsigned pg_num; /* Reference number */
unsigned nr_pgpaths; /* Number of paths in PG */
struct list_head pgpaths;
bool bypassed:1; /* Temporarily bypass this PG? */
};
/* Multipath context */
struct multipath {
unsigned long flags; /* Multipath state flags */
spinlock_t lock;
enum dm_queue_mode queue_mode;
struct pgpath *current_pgpath;
struct priority_group *current_pg;
struct priority_group *next_pg; /* Switch to this PG if set */
atomic_t nr_valid_paths; /* Total number of usable paths */
unsigned nr_priority_groups;
struct list_head priority_groups;
const char *hw_handler_name;
char *hw_handler_params;
wait_queue_head_t pg_init_wait; /* Wait for pg_init completion */
unsigned pg_init_retries; /* Number of times to retry pg_init */
unsigned pg_init_delay_msecs; /* Number of msecs before pg_init retry */
atomic_t pg_init_in_progress; /* Only one pg_init allowed at once */
atomic_t pg_init_count; /* Number of times pg_init called */
struct mutex work_mutex;
struct work_struct trigger_event;
struct dm_target *ti;
struct work_struct process_queued_bios;
struct bio_list queued_bios;
struct timer_list nopath_timer; /* Timeout for queue_if_no_path */
};
/*
* Context information attached to each io we process.
*/
struct dm_mpath_io {
struct pgpath *pgpath;
size_t nr_bytes;
u64 start_time_ns;
};
typedef int (*action_fn) (struct pgpath *pgpath);
static struct workqueue_struct *kmultipathd, *kmpath_handlerd;
static void trigger_event(struct work_struct *work);
static void activate_or_offline_path(struct pgpath *pgpath);
static void activate_path_work(struct work_struct *work);
static void process_queued_bios(struct work_struct *work);
static void queue_if_no_path_timeout_work(struct timer_list *t);
/*-----------------------------------------------
* Multipath state flags.
*-----------------------------------------------*/
#define MPATHF_QUEUE_IO 0 /* Must we queue all I/O? */
#define MPATHF_QUEUE_IF_NO_PATH 1 /* Queue I/O if last path fails? */
#define MPATHF_SAVED_QUEUE_IF_NO_PATH 2 /* Saved state during suspension */
#define MPATHF_RETAIN_ATTACHED_HW_HANDLER 3 /* If there's already a hw_handler present, don't change it. */
#define MPATHF_PG_INIT_DISABLED 4 /* pg_init is not currently allowed */
#define MPATHF_PG_INIT_REQUIRED 5 /* pg_init needs calling? */
#define MPATHF_PG_INIT_DELAY_RETRY 6 /* Delay pg_init retry? */
static bool mpath_double_check_test_bit(int MPATHF_bit, struct multipath *m)
{
bool r = test_bit(MPATHF_bit, &m->flags);
if (r) {
unsigned long flags;
spin_lock_irqsave(&m->lock, flags);
r = test_bit(MPATHF_bit, &m->flags);
spin_unlock_irqrestore(&m->lock, flags);
}
return r;
}
/*-----------------------------------------------
* Allocation routines
*-----------------------------------------------*/
static struct pgpath *alloc_pgpath(void)
{
struct pgpath *pgpath = kzalloc(sizeof(*pgpath), GFP_KERNEL);
if (!pgpath)
return NULL;
pgpath->is_active = true;
return pgpath;
}
static void free_pgpath(struct pgpath *pgpath)
{
kfree(pgpath);
}
static struct priority_group *alloc_priority_group(void)
{
struct priority_group *pg;
pg = kzalloc(sizeof(*pg), GFP_KERNEL);
if (pg)
INIT_LIST_HEAD(&pg->pgpaths);
return pg;
}
static void free_pgpaths(struct list_head *pgpaths, struct dm_target *ti)
{
struct pgpath *pgpath, *tmp;
list_for_each_entry_safe(pgpath, tmp, pgpaths, list) {
list_del(&pgpath->list);
dm_put_device(ti, pgpath->path.dev);
free_pgpath(pgpath);
}
}
static void free_priority_group(struct priority_group *pg,
struct dm_target *ti)
{
struct path_selector *ps = &pg->ps;
if (ps->type) {
ps->type->destroy(ps);
dm_put_path_selector(ps->type);
}
free_pgpaths(&pg->pgpaths, ti);
kfree(pg);
}
static struct multipath *alloc_multipath(struct dm_target *ti)
{
struct multipath *m;
m = kzalloc(sizeof(*m), GFP_KERNEL);
if (m) {
INIT_LIST_HEAD(&m->priority_groups);
spin_lock_init(&m->lock);
atomic_set(&m->nr_valid_paths, 0);
INIT_WORK(&m->trigger_event, trigger_event);
mutex_init(&m->work_mutex);
m->queue_mode = DM_TYPE_NONE;
m->ti = ti;
ti->private = m;
timer_setup(&m->nopath_timer, queue_if_no_path_timeout_work, 0);
}
return m;
}
static int alloc_multipath_stage2(struct dm_target *ti, struct multipath *m)
{
if (m->queue_mode == DM_TYPE_NONE) {
m->queue_mode = DM_TYPE_REQUEST_BASED;
} else if (m->queue_mode == DM_TYPE_BIO_BASED) {
INIT_WORK(&m->process_queued_bios, process_queued_bios);
/*
* bio-based doesn't support any direct scsi_dh management;
* it just discovers if a scsi_dh is attached.
*/
set_bit(MPATHF_RETAIN_ATTACHED_HW_HANDLER, &m->flags);
}
dm_table_set_type(ti->table, m->queue_mode);
/*
* Init fields that are only used when a scsi_dh is attached
* - must do this unconditionally (really doesn't hurt non-SCSI uses)
*/
set_bit(MPATHF_QUEUE_IO, &m->flags);
atomic_set(&m->pg_init_in_progress, 0);
atomic_set(&m->pg_init_count, 0);
m->pg_init_delay_msecs = DM_PG_INIT_DELAY_DEFAULT;
init_waitqueue_head(&m->pg_init_wait);
return 0;
}
static void free_multipath(struct multipath *m)
{
struct priority_group *pg, *tmp;
list_for_each_entry_safe(pg, tmp, &m->priority_groups, list) {
list_del(&pg->list);
free_priority_group(pg, m->ti);
}
kfree(m->hw_handler_name);
kfree(m->hw_handler_params);
mutex_destroy(&m->work_mutex);
kfree(m);
}
static struct dm_mpath_io *get_mpio(union map_info *info)
{
return info->ptr;
}
static size_t multipath_per_bio_data_size(void)
{
return sizeof(struct dm_mpath_io) + sizeof(struct dm_bio_details);
}
static struct dm_mpath_io *get_mpio_from_bio(struct bio *bio)
{
return dm_per_bio_data(bio, multipath_per_bio_data_size());
}
static struct dm_bio_details *get_bio_details_from_mpio(struct dm_mpath_io *mpio)
{
/* dm_bio_details is immediately after the dm_mpath_io in bio's per-bio-data */
void *bio_details = mpio + 1;
return bio_details;
}
static void multipath_init_per_bio_data(struct bio *bio, struct dm_mpath_io **mpio_p)
{
struct dm_mpath_io *mpio = get_mpio_from_bio(bio);
struct dm_bio_details *bio_details = get_bio_details_from_mpio(mpio);
mpio->nr_bytes = bio->bi_iter.bi_size;
mpio->pgpath = NULL;
mpio->start_time_ns = 0;
*mpio_p = mpio;
dm_bio_record(bio_details, bio);
}
/*-----------------------------------------------
* Path selection
*-----------------------------------------------*/
static int __pg_init_all_paths(struct multipath *m)
{
struct pgpath *pgpath;
unsigned long pg_init_delay = 0;
lockdep_assert_held(&m->lock);
if (atomic_read(&m->pg_init_in_progress) || test_bit(MPATHF_PG_INIT_DISABLED, &m->flags))
return 0;
atomic_inc(&m->pg_init_count);
clear_bit(MPATHF_PG_INIT_REQUIRED, &m->flags);
/* Check here to reset pg_init_required */
if (!m->current_pg)
return 0;
if (test_bit(MPATHF_PG_INIT_DELAY_RETRY, &m->flags))
pg_init_delay = msecs_to_jiffies(m->pg_init_delay_msecs != DM_PG_INIT_DELAY_DEFAULT ?
m->pg_init_delay_msecs : DM_PG_INIT_DELAY_MSECS);
list_for_each_entry(pgpath, &m->current_pg->pgpaths, list) {
/* Skip failed paths */
if (!pgpath->is_active)
continue;
if (queue_delayed_work(kmpath_handlerd, &pgpath->activate_path,
pg_init_delay))
atomic_inc(&m->pg_init_in_progress);
}
return atomic_read(&m->pg_init_in_progress);
}
static int pg_init_all_paths(struct multipath *m)
{
int ret;
unsigned long flags;
spin_lock_irqsave(&m->lock, flags);
ret = __pg_init_all_paths(m);
spin_unlock_irqrestore(&m->lock, flags);
return ret;
}
static void __switch_pg(struct multipath *m, struct priority_group *pg)
{
lockdep_assert_held(&m->lock);
m->current_pg = pg;
/* Must we initialise the PG first, and queue I/O till it's ready? */
if (m->hw_handler_name) {
set_bit(MPATHF_PG_INIT_REQUIRED, &m->flags);
set_bit(MPATHF_QUEUE_IO, &m->flags);
} else {
clear_bit(MPATHF_PG_INIT_REQUIRED, &m->flags);
clear_bit(MPATHF_QUEUE_IO, &m->flags);
}
atomic_set(&m->pg_init_count, 0);
}
static struct pgpath *choose_path_in_pg(struct multipath *m,
struct priority_group *pg,
size_t nr_bytes)
{
unsigned long flags;
struct dm_path *path;
struct pgpath *pgpath;
path = pg->ps.type->select_path(&pg->ps, nr_bytes);
if (!path)
return ERR_PTR(-ENXIO);
pgpath = path_to_pgpath(path);
if (unlikely(READ_ONCE(m->current_pg) != pg)) {
/* Only update current_pgpath if pg changed */
spin_lock_irqsave(&m->lock, flags);
m->current_pgpath = pgpath;
__switch_pg(m, pg);
spin_unlock_irqrestore(&m->lock, flags);
}
return pgpath;
}
static struct pgpath *choose_pgpath(struct multipath *m, size_t nr_bytes)
{
unsigned long flags;
struct priority_group *pg;
struct pgpath *pgpath;
unsigned bypassed = 1;
if (!atomic_read(&m->nr_valid_paths)) {
spin_lock_irqsave(&m->lock, flags);
clear_bit(MPATHF_QUEUE_IO, &m->flags);
spin_unlock_irqrestore(&m->lock, flags);
goto failed;
}
/* Were we instructed to switch PG? */
if (READ_ONCE(m->next_pg)) {
spin_lock_irqsave(&m->lock, flags);
pg = m->next_pg;
if (!pg) {
spin_unlock_irqrestore(&m->lock, flags);
goto check_current_pg;
}
m->next_pg = NULL;
spin_unlock_irqrestore(&m->lock, flags);
pgpath = choose_path_in_pg(m, pg, nr_bytes);
if (!IS_ERR_OR_NULL(pgpath))
return pgpath;
}
/* Don't change PG until it has no remaining paths */
check_current_pg:
pg = READ_ONCE(m->current_pg);
if (pg) {
pgpath = choose_path_in_pg(m, pg, nr_bytes);
if (!IS_ERR_OR_NULL(pgpath))
return pgpath;
}
/*
* Loop through priority groups until we find a valid path.
* First time we skip PGs marked 'bypassed'.
* Second time we only try the ones we skipped, but set
* pg_init_delay_retry so we do not hammer controllers.
*/
do {
list_for_each_entry(pg, &m->priority_groups, list) {
if (pg->bypassed == !!bypassed)
continue;
pgpath = choose_path_in_pg(m, pg, nr_bytes);
if (!IS_ERR_OR_NULL(pgpath)) {
if (!bypassed) {
spin_lock_irqsave(&m->lock, flags);
set_bit(MPATHF_PG_INIT_DELAY_RETRY, &m->flags);
spin_unlock_irqrestore(&m->lock, flags);
}
return pgpath;
}
}
} while (bypassed--);
failed:
spin_lock_irqsave(&m->lock, flags);
m->current_pgpath = NULL;
m->current_pg = NULL;
spin_unlock_irqrestore(&m->lock, flags);
return NULL;
}
/*
* dm_report_EIO() is a macro instead of a function to make pr_debug_ratelimited()
* report the function name and line number of the function from which
* it has been invoked.
*/
#define dm_report_EIO(m) \
do { \
DMDEBUG_LIMIT("%s: returning EIO; QIFNP = %d; SQIFNP = %d; DNFS = %d", \
dm_table_device_name((m)->ti->table), \
test_bit(MPATHF_QUEUE_IF_NO_PATH, &(m)->flags), \
test_bit(MPATHF_SAVED_QUEUE_IF_NO_PATH, &(m)->flags), \
dm_noflush_suspending((m)->ti)); \
} while (0)
/*
* Check whether bios must be queued in the device-mapper core rather
* than here in the target.
*/
static bool __must_push_back(struct multipath *m)
{
return dm_noflush_suspending(m->ti);
}
static bool must_push_back_rq(struct multipath *m)
{
unsigned long flags;
bool ret;
spin_lock_irqsave(&m->lock, flags);
ret = (test_bit(MPATHF_QUEUE_IF_NO_PATH, &m->flags) || __must_push_back(m));
spin_unlock_irqrestore(&m->lock, flags);
return ret;
}
/*
* Map cloned requests (request-based multipath)
*/
static int multipath_clone_and_map(struct dm_target *ti, struct request *rq,
union map_info *map_context,
struct request **__clone)
{
struct multipath *m = ti->private;
size_t nr_bytes = blk_rq_bytes(rq);
struct pgpath *pgpath;
struct block_device *bdev;
struct dm_mpath_io *mpio = get_mpio(map_context);
struct request_queue *q;
struct request *clone;
/* Do we need to select a new pgpath? */
pgpath = READ_ONCE(m->current_pgpath);
if (!pgpath || !mpath_double_check_test_bit(MPATHF_QUEUE_IO, m))
pgpath = choose_pgpath(m, nr_bytes);
if (!pgpath) {
if (must_push_back_rq(m))
return DM_MAPIO_DELAY_REQUEUE;
dm_report_EIO(m); /* Failed */
return DM_MAPIO_KILL;
} else if (mpath_double_check_test_bit(MPATHF_QUEUE_IO, m) ||
mpath_double_check_test_bit(MPATHF_PG_INIT_REQUIRED, m)) {
pg_init_all_paths(m);
return DM_MAPIO_DELAY_REQUEUE;
}
mpio->pgpath = pgpath;
mpio->nr_bytes = nr_bytes;
bdev = pgpath->path.dev->bdev;
q = bdev_get_queue(bdev);
clone = blk_mq_alloc_request(q, rq->cmd_flags | REQ_NOMERGE,
BLK_MQ_REQ_NOWAIT);
if (IS_ERR(clone)) {
/* EBUSY, ENODEV or EWOULDBLOCK: requeue */
if (blk_queue_dying(q)) {
atomic_inc(&m->pg_init_in_progress);
activate_or_offline_path(pgpath);
return DM_MAPIO_DELAY_REQUEUE;
}
/*
* blk-mq's SCHED_RESTART can cover this requeue, so we
* needn't deal with it by DELAY_REQUEUE. More importantly,
* we have to return DM_MAPIO_REQUEUE so that blk-mq can
* get the queue busy feedback (via BLK_STS_RESOURCE),
* otherwise I/O merging can suffer.
*/
return DM_MAPIO_REQUEUE;
}
clone->bio = clone->biotail = NULL;
clone->cmd_flags |= REQ_FAILFAST_TRANSPORT;
*__clone = clone;
if (pgpath->pg->ps.type->start_io)
pgpath->pg->ps.type->start_io(&pgpath->pg->ps,
&pgpath->path,
nr_bytes);
return DM_MAPIO_REMAPPED;
}
static void multipath_release_clone(struct request *clone,
union map_info *map_context)
{
if (unlikely(map_context)) {
/*
* non-NULL map_context means caller is still map
* method; must undo multipath_clone_and_map()
*/
struct dm_mpath_io *mpio = get_mpio(map_context);
struct pgpath *pgpath = mpio->pgpath;
if (pgpath && pgpath->pg->ps.type->end_io)
pgpath->pg->ps.type->end_io(&pgpath->pg->ps,
&pgpath->path,
mpio->nr_bytes,
clone->io_start_time_ns);
}
blk_mq_free_request(clone);
}
/*
* Map cloned bios (bio-based multipath)
*/
static void __multipath_queue_bio(struct multipath *m, struct bio *bio)
{
/* Queue for the daemon to resubmit */
bio_list_add(&m->queued_bios, bio);
if (!test_bit(MPATHF_QUEUE_IO, &m->flags))
queue_work(kmultipathd, &m->process_queued_bios);
}
static void multipath_queue_bio(struct multipath *m, struct bio *bio)
{
unsigned long flags;
spin_lock_irqsave(&m->lock, flags);
__multipath_queue_bio(m, bio);
spin_unlock_irqrestore(&m->lock, flags);
}
static struct pgpath *__map_bio(struct multipath *m, struct bio *bio)
{
struct pgpath *pgpath;
unsigned long flags;
/* Do we need to select a new pgpath? */
pgpath = READ_ONCE(m->current_pgpath);
if (!pgpath || !mpath_double_check_test_bit(MPATHF_QUEUE_IO, m))
pgpath = choose_pgpath(m, bio->bi_iter.bi_size);
if (!pgpath) {
spin_lock_irqsave(&m->lock, flags);
if (test_bit(MPATHF_QUEUE_IF_NO_PATH, &m->flags)) {
__multipath_queue_bio(m, bio);
pgpath = ERR_PTR(-EAGAIN);
}
spin_unlock_irqrestore(&m->lock, flags);
} else if (mpath_double_check_test_bit(MPATHF_QUEUE_IO, m) ||
mpath_double_check_test_bit(MPATHF_PG_INIT_REQUIRED, m)) {
multipath_queue_bio(m, bio);
pg_init_all_paths(m);
return ERR_PTR(-EAGAIN);
}
return pgpath;
}
static int __multipath_map_bio(struct multipath *m, struct bio *bio,
struct dm_mpath_io *mpio)
{
struct pgpath *pgpath = __map_bio(m, bio);
if (IS_ERR(pgpath))
return DM_MAPIO_SUBMITTED;
if (!pgpath) {
if (__must_push_back(m))
return DM_MAPIO_REQUEUE;
dm_report_EIO(m);
return DM_MAPIO_KILL;
}
mpio->pgpath = pgpath;
if (dm_ps_use_hr_timer(pgpath->pg->ps.type))
mpio->start_time_ns = ktime_get_ns();
bio->bi_status = 0;
bio_set_dev(bio, pgpath->path.dev->bdev);
bio->bi_opf |= REQ_FAILFAST_TRANSPORT;
if (pgpath->pg->ps.type->start_io)
pgpath->pg->ps.type->start_io(&pgpath->pg->ps,
&pgpath->path,
mpio->nr_bytes);
return DM_MAPIO_REMAPPED;
}
static int multipath_map_bio(struct dm_target *ti, struct bio *bio)
{
struct multipath *m = ti->private;
struct dm_mpath_io *mpio = NULL;
multipath_init_per_bio_data(bio, &mpio);
return __multipath_map_bio(m, bio, mpio);
}
static void process_queued_io_list(struct multipath *m)
{
if (m->queue_mode == DM_TYPE_REQUEST_BASED)
dm_mq_kick_requeue_list(dm_table_get_md(m->ti->table));
else if (m->queue_mode == DM_TYPE_BIO_BASED)
queue_work(kmultipathd, &m->process_queued_bios);
}
static void process_queued_bios(struct work_struct *work)
{
int r;
unsigned long flags;
struct bio *bio;
struct bio_list bios;
struct blk_plug plug;
struct multipath *m =
container_of(work, struct multipath, process_queued_bios);
bio_list_init(&bios);
spin_lock_irqsave(&m->lock, flags);
if (bio_list_empty(&m->queued_bios)) {
spin_unlock_irqrestore(&m->lock, flags);
return;
}
bio_list_merge(&bios, &m->queued_bios);
bio_list_init(&m->queued_bios);
spin_unlock_irqrestore(&m->lock, flags);
blk_start_plug(&plug);
while ((bio = bio_list_pop(&bios))) {
struct dm_mpath_io *mpio = get_mpio_from_bio(bio);
dm_bio_restore(get_bio_details_from_mpio(mpio), bio);
r = __multipath_map_bio(m, bio, mpio);
switch (r) {
case DM_MAPIO_KILL:
bio->bi_status = BLK_STS_IOERR;
bio_endio(bio);
break;
case DM_MAPIO_REQUEUE:
bio->bi_status = BLK_STS_DM_REQUEUE;
bio_endio(bio);
break;
case DM_MAPIO_REMAPPED:
submit_bio_noacct(bio);
break;
case DM_MAPIO_SUBMITTED:
break;
default:
WARN_ONCE(true, "__multipath_map_bio() returned %d\n", r);
}
}
blk_finish_plug(&plug);
}
/*
* If we run out of usable paths, should we queue I/O or error it?
*/
static int queue_if_no_path(struct multipath *m, bool queue_if_no_path,
bool save_old_value, const char *caller)
{
unsigned long flags;
bool queue_if_no_path_bit, saved_queue_if_no_path_bit;
const char *dm_dev_name = dm_table_device_name(m->ti->table);
DMDEBUG("%s: %s caller=%s queue_if_no_path=%d save_old_value=%d",
dm_dev_name, __func__, caller, queue_if_no_path, save_old_value);
spin_lock_irqsave(&m->lock, flags);
queue_if_no_path_bit = test_bit(MPATHF_QUEUE_IF_NO_PATH, &m->flags);
saved_queue_if_no_path_bit = test_bit(MPATHF_SAVED_QUEUE_IF_NO_PATH, &m->flags);
if (save_old_value) {
if (unlikely(!queue_if_no_path_bit && saved_queue_if_no_path_bit)) {
DMERR("%s: QIFNP disabled but saved as enabled, saving again loses state, not saving!",
dm_dev_name);
} else
assign_bit(MPATHF_SAVED_QUEUE_IF_NO_PATH, &m->flags, queue_if_no_path_bit);
} else if (!queue_if_no_path && saved_queue_if_no_path_bit) {
/* due to "fail_if_no_path" message, need to honor it. */
clear_bit(MPATHF_SAVED_QUEUE_IF_NO_PATH, &m->flags);
}
assign_bit(MPATHF_QUEUE_IF_NO_PATH, &m->flags, queue_if_no_path);
DMDEBUG("%s: after %s changes; QIFNP = %d; SQIFNP = %d; DNFS = %d",
dm_dev_name, __func__,
test_bit(MPATHF_QUEUE_IF_NO_PATH, &m->flags),
test_bit(MPATHF_SAVED_QUEUE_IF_NO_PATH, &m->flags),
dm_noflush_suspending(m->ti));
spin_unlock_irqrestore(&m->lock, flags);
if (!queue_if_no_path) {
dm_table_run_md_queue_async(m->ti->table);
process_queued_io_list(m);
}
return 0;
}
/*
* If the queue_if_no_path timeout fires, turn off queue_if_no_path and
* process any queued I/O.
*/
static void queue_if_no_path_timeout_work(struct timer_list *t)
{
struct multipath *m = from_timer(m, t, nopath_timer);
DMWARN("queue_if_no_path timeout on %s, failing queued IO",
dm_table_device_name(m->ti->table));
queue_if_no_path(m, false, false, __func__);
}
/*
* Enable the queue_if_no_path timeout if necessary.
* Called with m->lock held.
*/
static void enable_nopath_timeout(struct multipath *m)
{
unsigned long queue_if_no_path_timeout =
READ_ONCE(queue_if_no_path_timeout_secs) * HZ;
lockdep_assert_held(&m->lock);
if (queue_if_no_path_timeout > 0 &&
atomic_read(&m->nr_valid_paths) == 0 &&
test_bit(MPATHF_QUEUE_IF_NO_PATH, &m->flags)) {
mod_timer(&m->nopath_timer,
jiffies + queue_if_no_path_timeout);
}
}
static void disable_nopath_timeout(struct multipath *m)
{
del_timer_sync(&m->nopath_timer);
}
/*
* An event is triggered whenever a path is taken out of use.
* Includes path failure and PG bypass.
*/
static void trigger_event(struct work_struct *work)
{
struct multipath *m =
container_of(work, struct multipath, trigger_event);
dm_table_event(m->ti->table);
}
/*-----------------------------------------------------------------
* Constructor/argument parsing:
* <#multipath feature args> [<arg>]*
* <#hw_handler args> [hw_handler [<arg>]*]
* <#priority groups>
* <initial priority group>
* [<selector> <#selector args> [<arg>]*
* <#paths> <#per-path selector args>
* [<path> [<arg>]* ]+ ]+
*---------------------------------------------------------------*/
static int parse_path_selector(struct dm_arg_set *as, struct priority_group *pg,
struct dm_target *ti)
{
int r;
struct path_selector_type *pst;
unsigned ps_argc;
static const struct dm_arg _args[] = {
{0, 1024, "invalid number of path selector args"},
};
pst = dm_get_path_selector(dm_shift_arg(as));
if (!pst) {
ti->error = "unknown path selector type";
return -EINVAL;
}
r = dm_read_arg_group(_args, as, &ps_argc, &ti->error);
if (r) {
dm_put_path_selector(pst);
return -EINVAL;
}
r = pst->create(&pg->ps, ps_argc, as->argv);
if (r) {
dm_put_path_selector(pst);
ti->error = "path selector constructor failed";
return r;
}
pg->ps.type = pst;
dm_consume_args(as, ps_argc);
return 0;
}
static int setup_scsi_dh(struct block_device *bdev, struct multipath *m,
const char **attached_handler_name, char **error)
{
struct request_queue *q = bdev_get_queue(bdev);
int r;
if (mpath_double_check_test_bit(MPATHF_RETAIN_ATTACHED_HW_HANDLER, m)) {
retain:
if (*attached_handler_name) {
/*
* Clear any hw_handler_params associated with a
* handler that isn't already attached.
*/
if (m->hw_handler_name && strcmp(*attached_handler_name, m->hw_handler_name)) {
kfree(m->hw_handler_params);
m->hw_handler_params = NULL;
}
/*
* Reset hw_handler_name to match the attached handler
*
* NB. This modifies the table line to show the actual
* handler instead of the original table passed in.
*/
kfree(m->hw_handler_name);
m->hw_handler_name = *attached_handler_name;
*attached_handler_name = NULL;
}
}
if (m->hw_handler_name) {
r = scsi_dh_attach(q, m->hw_handler_name);
if (r == -EBUSY) {
DMINFO("retaining handler on device %pg", bdev);
goto retain;
}
if (r < 0) {
*error = "error attaching hardware handler";
return r;
}
if (m->hw_handler_params) {
r = scsi_dh_set_params(q, m->hw_handler_params);
if (r < 0) {
*error = "unable to set hardware handler parameters";
return r;
}
}
}
return 0;
}
static struct pgpath *parse_path(struct dm_arg_set *as, struct path_selector *ps,
struct dm_target *ti)
{
int r;
struct pgpath *p;
struct multipath *m = ti->private;
struct request_queue *q;
const char *attached_handler_name = NULL;
/* we need at least a path arg */
if (as->argc < 1) {
ti->error = "no device given";
return ERR_PTR(-EINVAL);
}
p = alloc_pgpath();
if (!p)
return ERR_PTR(-ENOMEM);
r = dm_get_device(ti, dm_shift_arg(as), dm_table_get_mode(ti->table),
&p->path.dev);
if (r) {
ti->error = "error getting device";
goto bad;
}
q = bdev_get_queue(p->path.dev->bdev);
attached_handler_name = scsi_dh_attached_handler_name(q, GFP_KERNEL);
if (attached_handler_name || m->hw_handler_name) {
INIT_DELAYED_WORK(&p->activate_path, activate_path_work);
r = setup_scsi_dh(p->path.dev->bdev, m, &attached_handler_name, &ti->error);
kfree(attached_handler_name);
if (r) {
dm_put_device(ti, p->path.dev);
goto bad;
}
}
r = ps->type->add_path(ps, &p->path, as->argc, as->argv, &ti->error);
if (r) {
dm_put_device(ti, p->path.dev);
goto bad;
}
return p;
bad:
free_pgpath(p);
return ERR_PTR(r);
}
static struct priority_group *parse_priority_group(struct dm_arg_set *as,
struct multipath *m)
{
static const struct dm_arg _args[] = {
{1, 1024, "invalid number of paths"},
{0, 1024, "invalid number of selector args"}
};
int r;
unsigned i, nr_selector_args, nr_args;
struct priority_group *pg;
struct dm_target *ti = m->ti;
if (as->argc < 2) {
as->argc = 0;
ti->error = "not enough priority group arguments";
return ERR_PTR(-EINVAL);
}
pg = alloc_priority_group();
if (!pg) {
ti->error = "couldn't allocate priority group";
return ERR_PTR(-ENOMEM);
}