aboutsummaryrefslogtreecommitdiff
path: root/drivers/block/rnbd/rnbd-clt.c
blob: 506e9094487f63c7376f5a0bd52fa6074d303d8d (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
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
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
// SPDX-License-Identifier: GPL-2.0-or-later
/*
 * RDMA Network Block Driver
 *
 * Copyright (c) 2014 - 2018 ProfitBricks GmbH. All rights reserved.
 * Copyright (c) 2018 - 2019 1&1 IONOS Cloud GmbH. All rights reserved.
 * Copyright (c) 2019 - 2020 1&1 IONOS SE. All rights reserved.
 */

#undef pr_fmt
#define pr_fmt(fmt) KBUILD_MODNAME " L" __stringify(__LINE__) ": " fmt

#include <linux/module.h>
#include <linux/blkdev.h>
#include <linux/hdreg.h>
#include <linux/scatterlist.h>
#include <linux/idr.h>

#include "rnbd-clt.h"

MODULE_DESCRIPTION("RDMA Network Block Device Client");
MODULE_LICENSE("GPL");

static int rnbd_client_major;
static DEFINE_IDA(index_ida);
static DEFINE_MUTEX(ida_lock);
static DEFINE_MUTEX(sess_lock);
static LIST_HEAD(sess_list);

/*
 * Maximum number of partitions an instance can have.
 * 6 bits = 64 minors = 63 partitions (one minor is used for the device itself)
 */
#define RNBD_PART_BITS		6

static inline bool rnbd_clt_get_sess(struct rnbd_clt_session *sess)
{
	return refcount_inc_not_zero(&sess->refcount);
}

static void free_sess(struct rnbd_clt_session *sess);

static void rnbd_clt_put_sess(struct rnbd_clt_session *sess)
{
	might_sleep();

	if (refcount_dec_and_test(&sess->refcount))
		free_sess(sess);
}

static void rnbd_clt_put_dev(struct rnbd_clt_dev *dev)
{
	might_sleep();

	if (!refcount_dec_and_test(&dev->refcount))
		return;

	mutex_lock(&ida_lock);
	ida_simple_remove(&index_ida, dev->clt_device_id);
	mutex_unlock(&ida_lock);
	kfree(dev->hw_queues);
	kfree(dev->pathname);
	rnbd_clt_put_sess(dev->sess);
	mutex_destroy(&dev->lock);
	kfree(dev);
}

static inline bool rnbd_clt_get_dev(struct rnbd_clt_dev *dev)
{
	return refcount_inc_not_zero(&dev->refcount);
}

static int rnbd_clt_set_dev_attr(struct rnbd_clt_dev *dev,
				 const struct rnbd_msg_open_rsp *rsp)
{
	struct rnbd_clt_session *sess = dev->sess;

	if (!rsp->logical_block_size)
		return -EINVAL;

	dev->device_id		    = le32_to_cpu(rsp->device_id);
	dev->nsectors		    = le64_to_cpu(rsp->nsectors);
	dev->logical_block_size	    = le16_to_cpu(rsp->logical_block_size);
	dev->physical_block_size    = le16_to_cpu(rsp->physical_block_size);
	dev->max_write_same_sectors = le32_to_cpu(rsp->max_write_same_sectors);
	dev->max_discard_sectors    = le32_to_cpu(rsp->max_discard_sectors);
	dev->discard_granularity    = le32_to_cpu(rsp->discard_granularity);
	dev->discard_alignment	    = le32_to_cpu(rsp->discard_alignment);
	dev->secure_discard	    = le16_to_cpu(rsp->secure_discard);
	dev->rotational		    = rsp->rotational;
	dev->wc 		    = !!(rsp->cache_policy & RNBD_WRITEBACK);
	dev->fua		    = !!(rsp->cache_policy & RNBD_FUA);

	dev->max_hw_sectors = sess->max_io_size / SECTOR_SIZE;
	dev->max_segments = BMAX_SEGMENTS;

	return 0;
}

static int rnbd_clt_change_capacity(struct rnbd_clt_dev *dev,
				    size_t new_nsectors)
{
	rnbd_clt_info(dev, "Device size changed from %zu to %zu sectors\n",
		       dev->nsectors, new_nsectors);
	dev->nsectors = new_nsectors;
	set_capacity_and_notify(dev->gd, dev->nsectors);
	return 0;
}

static int process_msg_open_rsp(struct rnbd_clt_dev *dev,
				struct rnbd_msg_open_rsp *rsp)
{
	int err = 0;

	mutex_lock(&dev->lock);
	if (dev->dev_state == DEV_STATE_UNMAPPED) {
		rnbd_clt_info(dev,
			       "Ignoring Open-Response message from server for  unmapped device\n");
		err = -ENOENT;
		goto out;
	}
	if (dev->dev_state == DEV_STATE_MAPPED_DISCONNECTED) {
		u64 nsectors = le64_to_cpu(rsp->nsectors);

		/*
		 * If the device was remapped and the size changed in the
		 * meantime we need to revalidate it
		 */
		if (dev->nsectors != nsectors)
			rnbd_clt_change_capacity(dev, nsectors);
		rnbd_clt_info(dev, "Device online, device remapped successfully\n");
	}
	err = rnbd_clt_set_dev_attr(dev, rsp);
	if (err)
		goto out;
	dev->dev_state = DEV_STATE_MAPPED;

out:
	mutex_unlock(&dev->lock);

	return err;
}

int rnbd_clt_resize_disk(struct rnbd_clt_dev *dev, size_t newsize)
{
	int ret = 0;

	mutex_lock(&dev->lock);
	if (dev->dev_state != DEV_STATE_MAPPED) {
		pr_err("Failed to set new size of the device, device is not opened\n");
		ret = -ENOENT;
		goto out;
	}
	ret = rnbd_clt_change_capacity(dev, newsize);

out:
	mutex_unlock(&dev->lock);

	return ret;
}

static inline void rnbd_clt_dev_requeue(struct rnbd_queue *q)
{
	if (WARN_ON(!q->hctx))
		return;

	/* We can come here from interrupt, thus async=true */
	blk_mq_run_hw_queue(q->hctx, true);
}

enum {
	RNBD_DELAY_IFBUSY = -1,
};

/**
 * rnbd_get_cpu_qlist() - finds a list with HW queues to be rerun
 * @sess:	Session to find a queue for
 * @cpu:	Cpu to start the search from
 *
 * Description:
 *     Each CPU has a list of HW queues, which needs to be rerun.  If a list
 *     is not empty - it is marked with a bit.  This function finds first
 *     set bit in a bitmap and returns corresponding CPU list.
 */
static struct rnbd_cpu_qlist *
rnbd_get_cpu_qlist(struct rnbd_clt_session *sess, int cpu)
{
	int bit;

	/* Search from cpu to nr_cpu_ids */
	bit = find_next_bit(sess->cpu_queues_bm, nr_cpu_ids, cpu);
	if (bit < nr_cpu_ids) {
		return per_cpu_ptr(sess->cpu_queues, bit);
	} else if (cpu != 0) {
		/* Search from 0 to cpu */
		bit = find_next_bit(sess->cpu_queues_bm, cpu, 0);
		if (bit < cpu)
			return per_cpu_ptr(sess->cpu_queues, bit);
	}

	return NULL;
}

static inline int nxt_cpu(int cpu)
{
	return (cpu + 1) % nr_cpu_ids;
}

/**
 * rnbd_rerun_if_needed() - rerun next queue marked as stopped
 * @sess:	Session to rerun a queue on
 *
 * Description:
 *     Each CPU has it's own list of HW queues, which should be rerun.
 *     Function finds such list with HW queues, takes a list lock, picks up
 *     the first HW queue out of the list and requeues it.
 *
 * Return:
 *     True if the queue was requeued, false otherwise.
 *
 * Context:
 *     Does not matter.
 */
static bool rnbd_rerun_if_needed(struct rnbd_clt_session *sess)
{
	struct rnbd_queue *q = NULL;
	struct rnbd_cpu_qlist *cpu_q;
	unsigned long flags;
	int *cpup;

	/*
	 * To keep fairness and not to let other queues starve we always
	 * try to wake up someone else in round-robin manner.  That of course
	 * increases latency but queues always have a chance to be executed.
	 */
	cpup = get_cpu_ptr(sess->cpu_rr);
	for (cpu_q = rnbd_get_cpu_qlist(sess, nxt_cpu(*cpup)); cpu_q;
	     cpu_q = rnbd_get_cpu_qlist(sess, nxt_cpu(cpu_q->cpu))) {
		if (!spin_trylock_irqsave(&cpu_q->requeue_lock, flags))
			continue;
		if (unlikely(!test_bit(cpu_q->cpu, sess->cpu_queues_bm)))
			goto unlock;
		q = list_first_entry_or_null(&cpu_q->requeue_list,
					     typeof(*q), requeue_list);
		if (WARN_ON(!q))
			goto clear_bit;
		list_del_init(&q->requeue_list);
		clear_bit_unlock(0, &q->in_list);

		if (list_empty(&cpu_q->requeue_list)) {
			/* Clear bit if nothing is left */
clear_bit:
			clear_bit(cpu_q->cpu, sess->cpu_queues_bm);
		}
unlock:
		spin_unlock_irqrestore(&cpu_q->requeue_lock, flags);

		if (q)
			break;
	}

	/**
	 * Saves the CPU that is going to be requeued on the per-cpu var. Just
	 * incrementing it doesn't work because rnbd_get_cpu_qlist() will
	 * always return the first CPU with something on the queue list when the
	 * value stored on the var is greater than the last CPU with something
	 * on the list.
	 */
	if (cpu_q)
		*cpup = cpu_q->cpu;
	put_cpu_var(sess->cpu_rr);

	if (q)
		rnbd_clt_dev_requeue(q);

	return q;
}

/**
 * rnbd_rerun_all_if_idle() - rerun all queues left in the list if
 *				 session is idling (there are no requests
 *				 in-flight).
 * @sess:	Session to rerun the queues on
 *
 * Description:
 *     This function tries to rerun all stopped queues if there are no
 *     requests in-flight anymore.  This function tries to solve an obvious
 *     problem, when number of tags < than number of queues (hctx), which
 *     are stopped and put to sleep.  If last permit, which has been just put,
 *     does not wake up all left queues (hctxs), IO requests hang forever.
 *
 *     That can happen when all number of permits, say N, have been exhausted
 *     from one CPU, and we have many block devices per session, say M.
 *     Each block device has it's own queue (hctx) for each CPU, so eventually
 *     we can put that number of queues (hctxs) to sleep: M x nr_cpu_ids.
 *     If number of permits N < M x nr_cpu_ids finally we will get an IO hang.
 *
 *     To avoid this hang last caller of rnbd_put_permit() (last caller is the
 *     one who observes sess->busy == 0) must wake up all remaining queues.
 *
 * Context:
 *     Does not matter.
 */
static void rnbd_rerun_all_if_idle(struct rnbd_clt_session *sess)
{
	bool requeued;

	do {
		requeued = rnbd_rerun_if_needed(sess);
	} while (atomic_read(&sess->busy) == 0 && requeued);
}

static struct rtrs_permit *rnbd_get_permit(struct rnbd_clt_session *sess,
					     enum rtrs_clt_con_type con_type,
					     int wait)
{
	struct rtrs_permit *permit;

	permit = rtrs_clt_get_permit(sess->rtrs, con_type,
				      wait ? RTRS_PERMIT_WAIT :
				      RTRS_PERMIT_NOWAIT);
	if (likely(permit))
		/* We have a subtle rare case here, when all permits can be
		 * consumed before busy counter increased.  This is safe,
		 * because loser will get NULL as a permit, observe 0 busy
		 * counter and immediately restart the queue himself.
		 */
		atomic_inc(&sess->busy);

	return permit;
}

static void rnbd_put_permit(struct rnbd_clt_session *sess,
			     struct rtrs_permit *permit)
{
	rtrs_clt_put_permit(sess->rtrs, permit);
	atomic_dec(&sess->busy);
	/* Paired with rnbd_clt_dev_add_to_requeue().  Decrement first
	 * and then check queue bits.
	 */
	smp_mb__after_atomic();
	rnbd_rerun_all_if_idle(sess);
}

static struct rnbd_iu *rnbd_get_iu(struct rnbd_clt_session *sess,
				     enum rtrs_clt_con_type con_type,
				     int wait)
{
	struct rnbd_iu *iu;
	struct rtrs_permit *permit;

	iu = kzalloc(sizeof(*iu), GFP_KERNEL);
	if (!iu) {
		return NULL;
	}

	permit = rnbd_get_permit(sess, con_type,
				  wait ? RTRS_PERMIT_WAIT :
				  RTRS_PERMIT_NOWAIT);
	if (unlikely(!permit)) {
		kfree(iu);
		return NULL;
	}

	iu->permit = permit;
	/*
	 * 1st reference is dropped after finishing sending a "user" message,
	 * 2nd reference is dropped after confirmation with the response is
	 * returned.
	 * 1st and 2nd can happen in any order, so the rnbd_iu should be
	 * released (rtrs_permit returned to rtrs) only after both
	 * are finished.
	 */
	atomic_set(&iu->refcount, 2);
	init_waitqueue_head(&iu->comp.wait);
	iu->comp.errno = INT_MAX;

	if (sg_alloc_table(&iu->sgt, 1, GFP_KERNEL)) {
		rnbd_put_permit(sess, permit);
		kfree(iu);
		return NULL;
	}

	return iu;
}

static void rnbd_put_iu(struct rnbd_clt_session *sess, struct rnbd_iu *iu)
{
	if (atomic_dec_and_test(&iu->refcount)) {
		sg_free_table(&iu->sgt);
		rnbd_put_permit(sess, iu->permit);
		kfree(iu);
	}
}

static void rnbd_softirq_done_fn(struct request *rq)
{
	struct rnbd_clt_dev *dev	= rq->rq_disk->private_data;
	struct rnbd_clt_session *sess	= dev->sess;
	struct rnbd_iu *iu;

	iu = blk_mq_rq_to_pdu(rq);
	sg_free_table_chained(&iu->sgt, RNBD_INLINE_SG_CNT);
	rnbd_put_permit(sess, iu->permit);
	blk_mq_end_request(rq, errno_to_blk_status(iu->errno));
}

static void msg_io_conf(void *priv, int errno)
{
	struct rnbd_iu *iu = priv;
	struct rnbd_clt_dev *dev = iu->dev;
	struct request *rq = iu->rq;
	int rw = rq_data_dir(rq);

	iu->errno = errno;

	blk_mq_complete_request(rq);

	if (errno)
		rnbd_clt_info_rl(dev, "%s I/O failed with err: %d\n",
				 rw == READ ? "read" : "write", errno);
}

static void wake_up_iu_comp(struct rnbd_iu *iu, int errno)
{
	iu->comp.errno = errno;
	wake_up(&iu->comp.wait);
}

static void msg_conf(void *priv, int errno)
{
	struct rnbd_iu *iu = priv;

	iu->errno = errno;
	schedule_work(&iu->work);
}

enum wait_type {
	NO_WAIT = 0,
	WAIT    = 1
};

static int send_usr_msg(struct rtrs_clt *rtrs, int dir,
			struct rnbd_iu *iu, struct kvec *vec,
			size_t len, struct scatterlist *sg, unsigned int sg_len,
			void (*conf)(struct work_struct *work),
			int *errno, enum wait_type wait)
{
	int err;
	struct rtrs_clt_req_ops req_ops;

	INIT_WORK(&iu->work, conf);
	req_ops = (struct rtrs_clt_req_ops) {
		.priv = iu,
		.conf_fn = msg_conf,
	};
	err = rtrs_clt_request(dir, &req_ops, rtrs, iu->permit,
				vec, 1, len, sg, sg_len);
	if (!err && wait) {
		wait_event(iu->comp.wait, iu->comp.errno != INT_MAX);
		*errno = iu->comp.errno;
	} else {
		*errno = 0;
	}

	return err;
}

static void msg_close_conf(struct work_struct *work)
{
	struct rnbd_iu *iu = container_of(work, struct rnbd_iu, work);
	struct rnbd_clt_dev *dev = iu->dev;

	wake_up_iu_comp(iu, iu->errno);
	rnbd_put_iu(dev->sess, iu);
	rnbd_clt_put_dev(dev);
}

static int send_msg_close(struct rnbd_clt_dev *dev, u32 device_id, bool wait)
{
	struct rnbd_clt_session *sess = dev->sess;
	struct rnbd_msg_close msg;
	struct rnbd_iu *iu;
	struct kvec vec = {
		.iov_base = &msg,
		.iov_len  = sizeof(msg)
	};
	int err, errno;

	iu = rnbd_get_iu(sess, RTRS_ADMIN_CON, RTRS_PERMIT_WAIT);
	if (!iu)
		return -ENOMEM;

	iu->buf = NULL;
	iu->dev = dev;

	msg.hdr.type	= cpu_to_le16(RNBD_MSG_CLOSE);
	msg.device_id	= cpu_to_le32(device_id);

	WARN_ON(!rnbd_clt_get_dev(dev));
	err = send_usr_msg(sess->rtrs, WRITE, iu, &vec, 0, NULL, 0,
			   msg_close_conf, &errno, wait);
	if (err) {
		rnbd_clt_put_dev(dev);
		rnbd_put_iu(sess, iu);
	} else {
		err = errno;
	}

	rnbd_put_iu(sess, iu);
	return err;
}

static void msg_open_conf(struct work_struct *work)
{
	struct rnbd_iu *iu = container_of(work, struct rnbd_iu, work);
	struct rnbd_msg_open_rsp *rsp = iu->buf;
	struct rnbd_clt_dev *dev = iu->dev;
	int errno = iu->errno;

	if (errno) {
		rnbd_clt_err(dev,
			      "Opening failed, server responded: %d\n",
			      errno);
	} else {
		errno = process_msg_open_rsp(dev, rsp);
		if (errno) {
			u32 device_id = le32_to_cpu(rsp->device_id);
			/*
			 * If server thinks its fine, but we fail to process
			 * then be nice and send a close to server.
			 */
			(void)send_msg_close(dev, device_id, NO_WAIT);
		}
	}
	kfree(rsp);
	wake_up_iu_comp(iu, errno);
	rnbd_put_iu(dev->sess, iu);
	rnbd_clt_put_dev(dev);
}

static void msg_sess_info_conf(struct work_struct *work)
{
	struct rnbd_iu *iu = container_of(work, struct rnbd_iu, work);
	struct rnbd_msg_sess_info_rsp *rsp = iu->buf;
	struct rnbd_clt_session *sess = iu->sess;

	if (!iu->errno)
		sess->ver = min_t(u8, rsp->ver, RNBD_PROTO_VER_MAJOR);

	kfree(rsp);
	wake_up_iu_comp(iu, iu->errno);
	rnbd_put_iu(sess, iu);
	rnbd_clt_put_sess(sess);
}

static int send_msg_open(struct rnbd_clt_dev *dev, bool wait)
{
	struct rnbd_clt_session *sess = dev->sess;
	struct rnbd_msg_open_rsp *rsp;
	struct rnbd_msg_open msg;
	struct rnbd_iu *iu;
	struct kvec vec = {
		.iov_base = &msg,
		.iov_len  = sizeof(msg)
	};
	int err, errno;

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

	iu = rnbd_get_iu(sess, RTRS_ADMIN_CON, RTRS_PERMIT_WAIT);
	if (!iu) {
		kfree(rsp);
		return -ENOMEM;
	}

	iu->buf = rsp;
	iu->dev = dev;

	sg_init_one(iu->sgt.sgl, rsp, sizeof(*rsp));

	msg.hdr.type	= cpu_to_le16(RNBD_MSG_OPEN);
	msg.access_mode	= dev->access_mode;
	strlcpy(msg.dev_name, dev->pathname, sizeof(msg.dev_name));

	WARN_ON(!rnbd_clt_get_dev(dev));
	err = send_usr_msg(sess->rtrs, READ, iu,
			   &vec, sizeof(*rsp), iu->sgt.sgl, 1,
			   msg_open_conf, &errno, wait);
	if (err) {
		rnbd_clt_put_dev(dev);
		rnbd_put_iu(sess, iu);
		kfree(rsp);
	} else {
		err = errno;
	}

	rnbd_put_iu(sess, iu);
	return err;
}

static int send_msg_sess_info(struct rnbd_clt_session *sess, bool wait)
{
	struct rnbd_msg_sess_info_rsp *rsp;
	struct rnbd_msg_sess_info msg;
	struct rnbd_iu *iu;
	struct kvec vec = {
		.iov_base = &msg,
		.iov_len  = sizeof(msg)
	};
	int err, errno;

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

	iu = rnbd_get_iu(sess, RTRS_ADMIN_CON, RTRS_PERMIT_WAIT);
	if (!iu) {
		kfree(rsp);
		return -ENOMEM;
	}

	iu->buf = rsp;
	iu->sess = sess;
	sg_init_one(iu->sgt.sgl, rsp, sizeof(*rsp));

	msg.hdr.type = cpu_to_le16(RNBD_MSG_SESS_INFO);
	msg.ver      = RNBD_PROTO_VER_MAJOR;

	if (!rnbd_clt_get_sess(sess)) {
		/*
		 * That can happen only in one case, when RTRS has restablished
		 * the connection and link_ev() is called, but session is almost
		 * dead, last reference on session is put and caller is waiting
		 * for RTRS to close everything.
		 */
		err = -ENODEV;
		goto put_iu;
	}
	err = send_usr_msg(sess->rtrs, READ, iu,
			   &vec, sizeof(*rsp), iu->sgt.sgl, 1,
			   msg_sess_info_conf, &errno, wait);
	if (err) {
		rnbd_clt_put_sess(sess);
put_iu:
		rnbd_put_iu(sess, iu);
		kfree(rsp);
	} else {
		err = errno;
	}
	rnbd_put_iu(sess, iu);
	return err;
}

static void set_dev_states_to_disconnected(struct rnbd_clt_session *sess)
{
	struct rnbd_clt_dev *dev;

	mutex_lock(&sess->lock);
	list_for_each_entry(dev, &sess->devs_list, list) {
		rnbd_clt_err(dev, "Device disconnected.\n");

		mutex_lock(&dev->lock);
		if (dev->dev_state == DEV_STATE_MAPPED)
			dev->dev_state = DEV_STATE_MAPPED_DISCONNECTED;
		mutex_unlock(&dev->lock);
	}
	mutex_unlock(&sess->lock);
}

static void remap_devs(struct rnbd_clt_session *sess)
{
	struct rnbd_clt_dev *dev;
	struct rtrs_attrs attrs;
	int err;

	/*
	 * Careful here: we are called from RTRS link event directly,
	 * thus we can't send any RTRS request and wait for response
	 * or RTRS will not be able to complete request with failure
	 * if something goes wrong (failing of outstanding requests
	 * happens exactly from the context where we are blocking now).
	 *
	 * So to avoid deadlocks each usr message sent from here must
	 * be asynchronous.
	 */

	err = send_msg_sess_info(sess, NO_WAIT);
	if (err) {
		pr_err("send_msg_sess_info(\"%s\"): %d\n", sess->sessname, err);
		return;
	}

	rtrs_clt_query(sess->rtrs, &attrs);
	mutex_lock(&sess->lock);
	sess->max_io_size = attrs.max_io_size;

	list_for_each_entry(dev, &sess->devs_list, list) {
		bool skip;

		mutex_lock(&dev->lock);
		skip = (dev->dev_state == DEV_STATE_INIT);
		mutex_unlock(&dev->lock);
		if (skip)
			/*
			 * When device is establishing connection for the first
			 * time - do not remap, it will be closed soon.
			 */
			continue;

		rnbd_clt_info(dev, "session reconnected, remapping device\n");
		err = send_msg_open(dev, NO_WAIT);
		if (err) {
			rnbd_clt_err(dev, "send_msg_open(): %d\n", err);
			break;
		}
	}
	mutex_unlock(&sess->lock);
}

static void rnbd_clt_link_ev(void *priv, enum rtrs_clt_link_ev ev)
{
	struct rnbd_clt_session *sess = priv;

	switch (ev) {
	case RTRS_CLT_LINK_EV_DISCONNECTED:
		set_dev_states_to_disconnected(sess);
		break;
	case RTRS_CLT_LINK_EV_RECONNECTED:
		remap_devs(sess);
		break;
	default:
		pr_err("Unknown session event received (%d), session: %s\n",
		       ev, sess->sessname);
	}
}

static void rnbd_init_cpu_qlists(struct rnbd_cpu_qlist __percpu *cpu_queues)
{
	unsigned int cpu;
	struct rnbd_cpu_qlist *cpu_q;

	for_each_possible_cpu(cpu) {
		cpu_q = per_cpu_ptr(cpu_queues, cpu);

		cpu_q->cpu = cpu;
		INIT_LIST_HEAD(&cpu_q->requeue_list);
		spin_lock_init(&cpu_q->requeue_lock);
	}
}

static void destroy_mq_tags(struct rnbd_clt_session *sess)
{
	if (sess->tag_set.tags)
		blk_mq_free_tag_set(&sess->tag_set);
}

static inline void wake_up_rtrs_waiters(struct rnbd_clt_session *sess)
{
	sess->rtrs_ready = true;
	wake_up_all(&sess->rtrs_waitq);
}

static void close_rtrs(struct rnbd_clt_session *sess)
{
	might_sleep();

	if (!IS_ERR_OR_NULL(sess->rtrs)) {
		rtrs_clt_close(sess->rtrs);
		sess->rtrs = NULL;
		wake_up_rtrs_waiters(sess);
	}
}

static void free_sess(struct rnbd_clt_session *sess)
{
	WARN_ON(!list_empty(&sess->devs_list));

	might_sleep();

	close_rtrs(sess);
	destroy_mq_tags(sess);
	if (!list_empty(&sess->list)) {
		mutex_lock(&sess_lock);
		list_del(&sess->list);
		mutex_unlock(&sess_lock);
	}
	free_percpu(sess->cpu_queues);
	free_percpu(sess->cpu_rr);
	mutex_destroy(&sess->lock);
	kfree(sess);
}

static struct rnbd_clt_session *alloc_sess(const char *sessname)
{
	struct rnbd_clt_session *sess;
	int err, cpu;

	sess = kzalloc_node(sizeof(*sess), GFP_KERNEL, NUMA_NO_NODE);
	if (!sess)
		return ERR_PTR(-ENOMEM);
	strlcpy(sess->sessname, sessname, sizeof(sess->sessname));
	atomic_set(&sess->busy, 0);
	mutex_init(&sess->lock);
	INIT_LIST_HEAD(&sess->devs_list);
	INIT_LIST_HEAD(&sess->list);
	bitmap_zero(sess->cpu_queues_bm, NR_CPUS);
	init_waitqueue_head(&sess->rtrs_waitq);
	refcount_set(&sess->refcount, 1);

	sess->cpu_queues = alloc_percpu(struct rnbd_cpu_qlist);
	if (!sess->cpu_queues) {
		err = -ENOMEM;
		goto err;
	}
	rnbd_init_cpu_qlists(sess->cpu_queues);

	/*
	 * That is simple percpu variable which stores cpu indices, which are
	 * incremented on each access.  We need that for the sake of fairness
	 * to wake up queues in a round-robin manner.
	 */
	sess->cpu_rr = alloc_percpu(int);
	if (!sess->cpu_rr) {
		err = -ENOMEM;
		goto err;
	}
	for_each_possible_cpu(cpu)
		* per_cpu_ptr(sess->cpu_rr, cpu) = cpu;

	return sess;

err:
	free_sess(sess);

	return ERR_PTR(err);
}

static int wait_for_rtrs_connection(struct rnbd_clt_session *sess)
{
	wait_event(sess->rtrs_waitq, sess->rtrs_ready);
	if (IS_ERR_OR_NULL(sess->rtrs))
		return -ECONNRESET;

	return 0;
}

static void wait_for_rtrs_disconnection(struct rnbd_clt_session *sess)
	__releases(&sess_lock)
	__acquires(&sess_lock)
{
	DEFINE_WAIT(wait);

	prepare_to_wait(&sess->rtrs_waitq, &wait, TASK_UNINTERRUPTIBLE);
	if (IS_ERR_OR_NULL(sess->rtrs)) {
		finish_wait(&sess->rtrs_waitq, &wait);
		return;
	}
	mutex_unlock(&sess_lock);
	/* loop in caller, see __find_and_get_sess().
	 * You can't leave mutex locked and call schedule(), you will catch a
	 * deadlock with a caller of free_sess(), which has just put the last
	 * reference and is about to take the sess_lock in order to delete
	 * the session from the list.
	 */
	schedule();
	mutex_lock(&sess_lock);
}

static struct rnbd_clt_session *__find_and_get_sess(const char *sessname)
	__releases(&sess_lock)
	__acquires(&sess_lock)
{
	struct rnbd_clt_session *sess, *sn;
	int err;

again:
	list_for_each_entry_safe(sess, sn, &sess_list, list) {
		if (strcmp(sessname, sess->sessname))
			continue;

		if (sess->rtrs_ready && IS_ERR_OR_NULL(sess->rtrs))
			/*
			 * No RTRS connection, session is dying.
			 */
			continue;

		if (rnbd_clt_get_sess(sess)) {
			/*
			 * Alive session is found, wait for RTRS connection.
			 */
			mutex_unlock(&sess_lock);
			err = wait_for_rtrs_connection(sess);
			if (err)
				rnbd_clt_put_sess(sess);
			mutex_lock(&sess_lock);

			if (err)
				/* Session is dying, repeat the loop */
				goto again;

			return sess;
		}
		/*
		 * Ref is 0, session is dying, wait for RTRS disconnect
		 * in order to avoid session names clashes.
		 */
		wait_for_rtrs_disconnection(sess);
		/*
		 * RTRS is disconnected and soon session will be freed,
		 * so repeat a loop.
		 */
		goto again;
	}

	return NULL;
}

static struct
rnbd_clt_session *find_or_create_sess(const char *sessname, bool *first)
{
	struct rnbd_clt_session *sess = NULL;

	mutex_lock(&sess_lock);
	sess = __find_and_get_sess(sessname);
	if (!sess) {
		sess = alloc_sess(sessname);
		if (IS_ERR(sess)) {
			mutex_unlock(&sess_lock);
			return sess;
		}
		list_add(&sess->list, &sess_list);
		*first = true;
	} else
		*first = false;
	mutex_unlock(&sess_lock);

	return sess;
}

static int rnbd_client_open(struct block_device *block_device, fmode_t mode)
{
	struct rnbd_clt_dev *dev = block_device->bd_disk->private_data;

	if (dev->read_only && (mode & FMODE_WRITE))
		return -EPERM;

	if (dev->dev_state == DEV_STATE_UNMAPPED ||
	    !rnbd_clt_get_dev(dev))
		return -EIO;

	return 0;
}

static void rnbd_client_release(struct gendisk *gen, fmode_t mode)
{
	struct rnbd_clt_dev *dev = gen->private_data;

	rnbd_clt_put_dev(dev);
}

static int rnbd_client_getgeo(struct block_device *block_device,
			      struct hd_geometry *geo)
{
	u64 size;
	struct rnbd_clt_dev *dev;

	dev = block_device->bd_disk->private_data;
	size = dev->size * (dev->logical_block_size / SECTOR_SIZE);
	geo->cylinders	= size >> 6;	/* size/64 */
	geo->heads	= 4;
	geo->sectors	= 16;
	geo->start	= 0;

	return 0;
}

static const struct block_device_operations rnbd_client_ops = {
	.owner		= THIS_MODULE,
	.open		= rnbd_client_open,
	.release	= rnbd_client_release,
	.getgeo		= rnbd_client_getgeo
};

/* The amount of data that belongs to an I/O and the amount of data that
 * should be read or written to the disk (bi_size) can differ.
 *
 * E.g. When WRITE_SAME is used, only a small amount of data is
 * transferred that is then written repeatedly over a lot of sectors.
 *
 * Get the size of data to be transferred via RTRS by summing up the size
 * of the scather-gather list entries.
 */
static size_t rnbd_clt_get_sg_size(struct scatterlist *sglist, u32 len)
{
	struct scatterlist *sg;
	size_t tsize = 0;
	int i;

	for_each_sg(sglist, sg, len, i)
		tsize += sg->length;
	return tsize;
}

static int rnbd_client_xfer_request(struct rnbd_clt_dev *dev,
				     struct request *rq,
				     struct rnbd_iu *iu)
{
	struct rtrs_clt *rtrs = dev->sess->rtrs;
	struct rtrs_permit *permit = iu->permit;
	struct rnbd_msg_io msg;
	struct rtrs_clt_req_ops req_ops;
	unsigned int sg_cnt = 0;
	struct kvec vec;
	size_t size;
	int err;

	iu->rq		= rq;
	iu->dev		= dev;
	msg.sector	= cpu_to_le64(blk_rq_pos(rq));
	msg.bi_size	= cpu_to_le32(blk_rq_bytes(rq));
	msg.rw		= cpu_to_le32(rq_to_rnbd_flags(rq));
	msg.prio	= cpu_to_le16(req_get_ioprio(rq));

	/*
	 * We only support discards with single segment for now.
	 * See queue limits.
	 */
	if (req_op(rq) != REQ_OP_DISCARD)
		sg_cnt = blk_rq_map_sg(dev->queue, rq, iu->sgt.sgl);

	if (sg_cnt == 0)
		sg_mark_end(&iu->sgt.sgl[0]);

	msg.hdr.type	= cpu_to_le16(RNBD_MSG_IO);
	msg.device_id	= cpu_to_le32(dev->device_id);

	vec = (struct kvec) {
		.iov_base = &msg,
		.iov_len  = sizeof(msg)
	};
	size = rnbd_clt_get_sg_size(iu->sgt.sgl, sg_cnt);
	req_ops = (struct rtrs_clt_req_ops) {
		.priv = iu,
		.conf_fn = msg_io_conf,
	};
	err = rtrs_clt_request(rq_data_dir(rq), &req_ops, rtrs, permit,
			       &vec, 1, size, iu->sgt.sgl, sg_cnt);
	if (unlikely(err)) {
		rnbd_clt_err_rl(dev, "RTRS failed to transfer IO, err: %d\n",
				 err);
		return err;
	}

	return 0;
}

/**
 * rnbd_clt_dev_add_to_requeue() - add device to requeue if session is busy
 * @dev:	Device to be checked
 * @q:		Queue to be added to the requeue list if required
 *
 * Description:
 *     If session is busy, that means someone will requeue us when resources
 *     are freed.  If session is not doing anything - device is not added to
 *     the list and @false is returned.
 */
static bool rnbd_clt_dev_add_to_requeue(struct rnbd_clt_dev *dev,
						struct rnbd_queue *q)
{
	struct rnbd_clt_session *sess = dev->sess;
	struct rnbd_cpu_qlist *cpu_q;
	unsigned long flags;
	bool added = true;
	bool need_set;

	cpu_q = get_cpu_ptr(sess->cpu_queues);
	spin_lock_irqsave(&cpu_q->requeue_lock, flags);

	if (likely(!test_and_set_bit_lock(0, &q->in_list))) {
		if (WARN_ON(!list_empty(&q->requeue_list)))
			goto unlock;

		need_set = !test_bit(cpu_q->cpu, sess->cpu_queues_bm);
		if (need_set) {
			set_bit(cpu_q->cpu, sess->cpu_queues_bm);
			/* Paired with rnbd_put_permit(). Set a bit first
			 * and then observe the busy counter.
			 */
			smp_mb__before_atomic();
		}
		if (likely(atomic_read(&sess->busy))) {
			list_add_tail(&q->requeue_list, &cpu_q->requeue_list);
		} else {
			/* Very unlikely, but possible: busy counter was
			 * observed as zero.  Drop all bits and return
			 * false to restart the queue by ourselves.
			 */
			if (need_set)
				clear_bit(cpu_q->cpu, sess->cpu_queues_bm);
			clear_bit_unlock(0, &q->in_list);
			added = false;
		}
	}
unlock:
	spin_unlock_irqrestore(&cpu_q->requeue_lock, flags);
	put_cpu_ptr(sess->cpu_queues);

	return added;
}

static void rnbd_clt_dev_kick_mq_queue(struct rnbd_clt_dev *dev,
					struct blk_mq_hw_ctx *hctx,
					int delay)
{
	struct rnbd_queue *q = hctx->driver_data;

	if (delay != RNBD_DELAY_IFBUSY)
		blk_mq_delay_run_hw_queue(hctx, delay);
	else if (unlikely(!rnbd_clt_dev_add_to_requeue(dev, q)))
		/*
		 * If session is not busy we have to restart
		 * the queue ourselves.
		 */
		blk_mq_delay_run_hw_queue(hctx, 10/*ms*/);
}

static blk_status_t rnbd_queue_rq(struct blk_mq_hw_ctx *hctx,
				   const struct blk_mq_queue_data *bd)
{
	struct request *rq = bd->rq;
	struct rnbd_clt_dev *dev = rq->rq_disk->private_data;
	struct rnbd_iu *iu = blk_mq_rq_to_pdu(rq);
	int err;
	blk_status_t ret = BLK_STS_IOERR;

	if (unlikely(dev->dev_state != DEV_STATE_MAPPED))
		return BLK_STS_IOERR;

	iu->permit = rnbd_get_permit(dev->sess, RTRS_IO_CON,
				      RTRS_PERMIT_NOWAIT);
	if (unlikely(!iu->permit)) {
		rnbd_clt_dev_kick_mq_queue(dev, hctx, RNBD_DELAY_IFBUSY);
		return BLK_STS_RESOURCE;
	}

	iu->sgt.sgl = iu->first_sgl;
	err = sg_alloc_table_chained(&iu->sgt,
				     /* Even-if the request has no segment,
				      * sglist must have one entry at least */
				     blk_rq_nr_phys_segments(rq) ? : 1,
				     iu->sgt.sgl,
				     RNBD_INLINE_SG_CNT);
	if (err) {
		rnbd_clt_err_rl(dev, "sg_alloc_table_chained ret=%d\n", err);
		rnbd_clt_dev_kick_mq_queue(dev, hctx, 10/*ms*/);
		rnbd_put_permit(dev->sess, iu->permit);
		return BLK_STS_RESOURCE;
	}

	blk_mq_start_request(rq);
	err = rnbd_client_xfer_request(dev, rq, iu);
	if (likely(err == 0))
		return BLK_STS_OK;
	if (unlikely(err == -EAGAIN || err == -ENOMEM)) {
		rnbd_clt_dev_kick_mq_queue(dev, hctx, 10/*ms*/);
		ret = BLK_STS_RESOURCE;
	}
	sg_free_table_chained(&iu->sgt, RNBD_INLINE_SG_CNT);
	rnbd_put_permit(dev->sess, iu->permit);
	return ret;
}

static struct blk_mq_ops rnbd_mq_ops = {
	.queue_rq	= rnbd_queue_rq,
	.complete	= rnbd_softirq_done_fn,
};

static int setup_mq_tags(struct rnbd_clt_session *sess)
{
	struct blk_mq_tag_set *tag_set = &sess->tag_set;

	memset(tag_set, 0, sizeof(*tag_set));
	tag_set->ops		= &rnbd_mq_ops;
	tag_set->queue_depth	= sess->queue_depth;
	tag_set->numa_node		= NUMA_NO_NODE;
	tag_set->flags		= BLK_MQ_F_SHOULD_MERGE |
				  BLK_MQ_F_TAG_QUEUE_SHARED;
	tag_set->cmd_size	= sizeof(struct rnbd_iu) + RNBD_RDMA_SGL_SIZE;
	tag_set->nr_hw_queues	= num_online_cpus();

	return blk_mq_alloc_tag_set(tag_set);
}

static struct rnbd_clt_session *
find_and_get_or_create_sess(const char *sessname,
			    const struct rtrs_addr *paths,
			    size_t path_cnt, u16 port_nr)
{
	struct rnbd_clt_session *sess;
	struct rtrs_attrs attrs;
	int err;
	bool first;
	struct rtrs_clt_ops rtrs_ops;

	sess = find_or_create_sess(sessname, &first);
	if (sess == ERR_PTR(-ENOMEM))
		return ERR_PTR(-ENOMEM);
	else if (!first)
		return sess;

	if (!path_cnt) {
		pr_err("Session %s not found, and path parameter not given", sessname);
		err = -ENXIO;
		goto put_sess;
	}

	rtrs_ops = (struct rtrs_clt_ops) {
		.priv = sess,
		.link_ev = rnbd_clt_link_ev,
	};
	/*
	 * Nothing was found, establish rtrs connection and proceed further.
	 */
	sess->rtrs = rtrs_clt_open(&rtrs_ops, sessname,
				   paths, path_cnt, port_nr,
				   0, /* Do not use pdu of rtrs */
				   RECONNECT_DELAY, BMAX_SEGMENTS,
				   BLK_MAX_SEGMENT_SIZE,
				   MAX_RECONNECTS);
	if (IS_ERR(sess->rtrs)) {
		err = PTR_ERR(sess->rtrs);
		goto wake_up_and_put;
	}
	rtrs_clt_query(sess->rtrs, &attrs);
	sess->max_io_size = attrs.max_io_size;
	sess->queue_depth = attrs.queue_depth;

	err = setup_mq_tags(sess);
	if (err)
		goto close_rtrs;

	err = send_msg_sess_info(sess, WAIT);
	if (err)
		goto close_rtrs;

	wake_up_rtrs_waiters(sess);

	return sess;

close_rtrs:
	close_rtrs(sess);
put_sess:
	rnbd_clt_put_sess(sess);

	return ERR_PTR(err);

wake_up_and_put:
	wake_up_rtrs_waiters(sess);
	goto put_sess;
}

static inline void rnbd_init_hw_queue(struct rnbd_clt_dev *dev,
				       struct rnbd_queue *q,
				       struct blk_mq_hw_ctx *hctx)
{
	INIT_LIST_HEAD(&q->requeue_list);
	q->dev  = dev;
	q->hctx = hctx;
}

static void rnbd_init_mq_hw_queues(struct rnbd_clt_dev *dev)
{
	int i;
	struct blk_mq_hw_ctx *hctx;
	struct rnbd_queue *q;

	queue_for_each_hw_ctx(dev->queue, hctx, i) {
		q = &dev->hw_queues[i];
		rnbd_init_hw_queue(dev, q, hctx);
		hctx->driver_data = q;
	}
}

static int setup_mq_dev(struct rnbd_clt_dev *dev)
{
	dev->queue = blk_mq_init_queue(&dev->sess->tag_set);
	if (IS_ERR(dev->queue)) {
		rnbd_clt_err(dev, "Initializing multiqueue queue failed, err: %ld\n",
			      PTR_ERR(dev->queue));
		return PTR_ERR(dev->queue);
	}
	rnbd_init_mq_hw_queues(dev);
	return 0;
}

static void setup_request_queue(struct rnbd_clt_dev *dev)
{
	blk_queue_logical_block_size(dev->queue, dev->logical_block_size);
	blk_queue_physical_block_size(dev->queue, dev->physical_block_size);
	blk_queue_max_hw_sectors(dev->queue, dev->max_hw_sectors);
	blk_queue_max_write_same_sectors(dev->queue,
					 dev->max_write_same_sectors);

	/*
	 * we don't support discards to "discontiguous" segments
	 * in on request
	 */
	blk_queue_max_discard_segments(dev->queue, 1);

	blk_queue_max_discard_sectors(dev->queue, dev->max_discard_sectors);
	dev->queue->limits.discard_granularity	= dev->discard_granularity;
	dev->queue->limits.discard_alignment	= dev->discard_alignment;
	if (dev->max_discard_sectors)
		blk_queue_flag_set(QUEUE_FLAG_DISCARD, dev->queue);
	if (dev->secure_discard)
		blk_queue_flag_set(QUEUE_FLAG_SECERASE, dev->queue);

	blk_queue_flag_set(QUEUE_FLAG_SAME_COMP, dev->queue);
	blk_queue_flag_set(QUEUE_FLAG_SAME_FORCE, dev->queue);
	blk_queue_max_segments(dev->queue, dev->max_segments);
	blk_queue_io_opt(dev->queue, dev->sess->max_io_size);
	blk_queue_virt_boundary(dev->queue, SZ_4K - 1);
	blk_queue_write_cache(dev->queue, dev->wc, dev->fua);
	dev->queue->queuedata = dev;
}

static void rnbd_clt_setup_gen_disk(struct rnbd_clt_dev *dev, int idx)
{
	dev->gd->major		= rnbd_client_major;
	dev->gd->first_minor	= idx << RNBD_PART_BITS;
	dev->gd->fops		= &rnbd_client_ops;
	dev->gd->queue		= dev->queue;
	dev->gd->private_data	= dev;
	snprintf(dev->gd->disk_name, sizeof(dev->gd->disk_name), "rnbd%d",
		 idx);
	pr_debug("disk_name=%s, capacity=%zu\n",
		 dev->gd->disk_name,
		 dev->nsectors * (dev->logical_block_size / SECTOR_SIZE)
		 );

	set_capacity(dev->gd, dev->nsectors);

	if (dev->access_mode == RNBD_ACCESS_RO) {
		dev->read_only = true;
		set_disk_ro(dev->gd, true);
	} else {
		dev->read_only = false;
	}

	if (!dev->rotational)
		blk_queue_flag_set(QUEUE_FLAG_NONROT, dev->queue);
}

static int rnbd_client_setup_device(struct rnbd_clt_session *sess,
				     struct rnbd_clt_dev *dev, int idx)
{
	int err;

	dev->size = dev->nsectors * dev->logical_block_size;

	err = setup_mq_dev(dev);
	if (err)
		return err;

	setup_request_queue(dev);

	dev->gd = alloc_disk_node(1 << RNBD_PART_BITS,	NUMA_NO_NODE);
	if (!dev->gd) {
		blk_cleanup_queue(dev->queue);
		return -ENOMEM;
	}

	rnbd_clt_setup_gen_disk(dev, idx);

	return 0;
}

static struct rnbd_clt_dev *init_dev(struct rnbd_clt_session *sess,
				      enum rnbd_access_mode access_mode,
				      const char *pathname)
{
	struct rnbd_clt_dev *dev;
	int ret;

	dev = kzalloc_node(sizeof(*dev), GFP_KERNEL, NUMA_NO_NODE);
	if (!dev)
		return ERR_PTR(-ENOMEM);

	dev->hw_queues = kcalloc(nr_cpu_ids, sizeof(*dev->hw_queues),
				 GFP_KERNEL);
	if (!dev->hw_queues) {
		ret = -ENOMEM;
		goto out_alloc;
	}

	mutex_lock(&ida_lock);
	ret = ida_simple_get(&index_ida, 0, 1 << (MINORBITS - RNBD_PART_BITS),
			     GFP_KERNEL);
	mutex_unlock(&ida_lock);
	if (ret < 0) {
		pr_err("Failed to initialize device '%s' from session %s, allocating idr failed, err: %d\n",
		       pathname, sess->sessname, ret);
		goto out_queues;
	}

	dev->pathname = kstrdup(pathname, GFP_KERNEL);
	if (!dev->pathname) {
		ret = -ENOMEM;
		goto out_queues;
	}

	dev->clt_device_id	= ret;
	dev->sess		= sess;
	dev->access_mode	= access_mode;
	mutex_init(&dev->lock);
	refcount_set(&dev->refcount, 1);
	dev->dev_state = DEV_STATE_INIT;

	/*
	 * Here we called from sysfs entry, thus clt-sysfs is
	 * responsible that session will not disappear.
	 */
	WARN_ON(!rnbd_clt_get_sess(sess));

	return dev;

out_queues:
	kfree(dev->hw_queues);
out_alloc:
	kfree(dev);
	return ERR_PTR(ret);
}

static bool __exists_dev(const char *pathname, const char *sessname)
{
	struct rnbd_clt_session *sess;
	struct rnbd_clt_dev *dev;
	bool found = false;

	list_for_each_entry(sess, &sess_list, list) {
		if (sessname && strncmp(sess->sessname, sessname,
					sizeof(sess->sessname)))
			continue;
		mutex_lock(&sess->lock);
		list_for_each_entry(dev, &sess->devs_list, list) {
			if (strlen(dev->pathname) == strlen(pathname) &&
			    !strcmp(dev->pathname, pathname)) {
				found = true;
				break;
			}
		}
		mutex_unlock(&sess->lock);
		if (found)
			break;
	}

	return found;
}

static bool exists_devpath(const char *pathname, const char *sessname)
{
	bool found;

	mutex_lock(&sess_lock);
	found = __exists_dev(pathname, sessname);
	mutex_unlock(&sess_lock);

	return found;
}

static bool insert_dev_if_not_exists_devpath(const char *pathname,
					     struct rnbd_clt_session *sess,
					     struct rnbd_clt_dev *dev)
{
	bool found;

	mutex_lock(&sess_lock);
	found = __exists_dev(pathname, sess->sessname);
	if (!found) {
		mutex_lock(&sess->lock);
		list_add_tail(&dev->list, &sess->devs_list);
		mutex_unlock(&sess->lock);
	}
	mutex_unlock(&sess_lock);

	return found;
}

static void delete_dev(struct rnbd_clt_dev *dev)
{
	struct rnbd_clt_session *sess = dev->sess;

	mutex_lock(&sess->lock);
	list_del(&dev->list);
	mutex_unlock(&sess->lock);
}

struct rnbd_clt_dev *rnbd_clt_map_device(const char *sessname,
					   struct rtrs_addr *paths,
					   size_t path_cnt, u16 port_nr,
					   const char *pathname,
					   enum rnbd_access_mode access_mode)
{
	struct rnbd_clt_session *sess;
	struct rnbd_clt_dev *dev;
	int ret;

	if (unlikely(exists_devpath(pathname, sessname)))
		return ERR_PTR(-EEXIST);

	sess = find_and_get_or_create_sess(sessname, paths, path_cnt, port_nr);
	if (IS_ERR(sess))
		return ERR_CAST(sess);

	dev = init_dev(sess, access_mode, pathname);
	if (IS_ERR(dev)) {
		pr_err("map_device: failed to map device '%s' from session %s, can't initialize device, err: %ld\n",
		       pathname, sess->sessname, PTR_ERR(dev));
		ret = PTR_ERR(dev);
		goto put_sess;
	}
	if (insert_dev_if_not_exists_devpath(pathname, sess, dev)) {
		ret = -EEXIST;
		goto put_dev;
	}
	ret = send_msg_open(dev, WAIT);
	if (ret) {
		rnbd_clt_err(dev,
			      "map_device: failed, can't open remote device, err: %d\n",
			      ret);
		goto del_dev;
	}
	mutex_lock(&dev->lock);
	pr_debug("Opened remote device: session=%s, path='%s'\n",
		 sess->sessname, pathname);
	ret = rnbd_client_setup_device(sess, dev, dev->clt_device_id);
	if (ret) {
		rnbd_clt_err(dev,
			      "map_device: Failed to configure device, err: %d\n",
			      ret);
		mutex_unlock(&dev->lock);
		goto send_close;
	}

	rnbd_clt_info(dev,
		       "map_device: Device mapped as %s (nsectors: %zu, logical_block_size: %d, physical_block_size: %d, max_write_same_sectors: %d, max_discard_sectors: %d, discard_granularity: %d, discard_alignment: %d, secure_discard: %d, max_segments: %d, max_hw_sectors: %d, rotational: %d, wc: %d, fua: %d)\n",
		       dev->gd->disk_name, dev->nsectors,
		       dev->logical_block_size, dev->physical_block_size,
		       dev->max_write_same_sectors, dev->max_discard_sectors,
		       dev->discard_granularity, dev->discard_alignment,
		       dev->secure_discard, dev->max_segments,
		       dev->max_hw_sectors, dev->rotational, dev->wc, dev->fua);

	mutex_unlock(&dev->lock);

	add_disk(dev->gd);
	rnbd_clt_put_sess(sess);

	return dev;

send_close:
	send_msg_close(dev, dev->device_id, WAIT);
del_dev:
	delete_dev(dev);
put_dev:
	rnbd_clt_put_dev(dev);
put_sess:
	rnbd_clt_put_sess(sess);

	return ERR_PTR(ret);
}

static void destroy_gen_disk(struct rnbd_clt_dev *dev)
{
	del_gendisk(dev->gd);
	blk_cleanup_queue(dev->queue);
	put_disk(dev->gd);
}

static void destroy_sysfs(struct rnbd_clt_dev *dev,
			  const struct attribute *sysfs_self)
{
	rnbd_clt_remove_dev_symlink(dev);
	if (dev->kobj.state_initialized) {
		if (sysfs_self)
			/* To avoid deadlock firstly remove itself */
			sysfs_remove_file_self(&dev->kobj, sysfs_self);
		kobject_del(&dev->kobj);
		kobject_put(&dev->kobj);
	}
}

int rnbd_clt_unmap_device(struct rnbd_clt_dev *dev, bool force,
			   const struct attribute *sysfs_self)
{
	struct rnbd_clt_session *sess = dev->sess;
	int refcount, ret = 0;
	bool was_mapped;

	mutex_lock(&dev->lock);
	if (dev->dev_state == DEV_STATE_UNMAPPED) {
		rnbd_clt_info(dev, "Device is already being unmapped\n");
		ret = -EALREADY;
		goto err;
	}
	refcount = refcount_read(&dev->refcount);
	if (!force && refcount > 1) {
		rnbd_clt_err(dev,
			      "Closing device failed, device is in use, (%d device users)\n",
			      refcount - 1);
		ret = -EBUSY;
		goto err;
	}
	was_mapped = (dev->dev_state == DEV_STATE_MAPPED);
	dev->dev_state = DEV_STATE_UNMAPPED;
	mutex_unlock(&dev->lock);

	delete_dev(dev);
	destroy_sysfs(dev, sysfs_self);
	destroy_gen_disk(dev);
	if (was_mapped && sess->rtrs)
		send_msg_close(dev, dev->device_id, WAIT);

	rnbd_clt_info(dev, "Device is unmapped\n");

	/* Likely last reference put */
	rnbd_clt_put_dev(dev);

	/*
	 * Here device and session can be vanished!
	 */

	return 0;
err:
	mutex_unlock(&dev->lock);

	return ret;
}

int rnbd_clt_remap_device(struct rnbd_clt_dev *dev)
{
	int err;

	mutex_lock(&dev->lock);
	if (dev->dev_state == DEV_STATE_MAPPED_DISCONNECTED)
		err = 0;
	else if (dev->dev_state == DEV_STATE_UNMAPPED)
		err = -ENODEV;
	else if (dev->dev_state == DEV_STATE_MAPPED)
		err = -EALREADY;
	else
		err = -EBUSY;
	mutex_unlock(&dev->lock);
	if (!err) {
		rnbd_clt_info(dev, "Remapping device.\n");
		err = send_msg_open(dev, WAIT);
		if (err)
			rnbd_clt_err(dev, "remap_device: %d\n", err);
	}

	return err;
}

static void unmap_device_work(struct work_struct *work)
{
	struct rnbd_clt_dev *dev;

	dev = container_of(work, typeof(*dev), unmap_on_rmmod_work);
	rnbd_clt_unmap_device(dev, true, NULL);
}

static void rnbd_destroy_sessions(void)
{
	struct rnbd_clt_session *sess, *sn;
	struct rnbd_clt_dev *dev, *tn;

	/* Firstly forbid access through sysfs interface */
	rnbd_clt_destroy_default_group();
	rnbd_clt_destroy_sysfs_files();

	/*
	 * Here at this point there is no any concurrent access to sessions
	 * list and devices list:
	 *   1. New session or device can't be created - session sysfs files
	 *      are removed.
	 *   2. Device or session can't be removed - module reference is taken
	 *      into account in unmap device sysfs callback.
	 *   3. No IO requests inflight - each file open of block_dev increases
	 *      module reference in get_disk().
	 *
	 * But still there can be user requests inflights, which are sent by
	 * asynchronous send_msg_*() functions, thus before unmapping devices
	 * RTRS session must be explicitly closed.
	 */

	list_for_each_entry_safe(sess, sn, &sess_list, list) {
		WARN_ON(!rnbd_clt_get_sess(sess));
		close_rtrs(sess);
		list_for_each_entry_safe(dev, tn, &sess->devs_list, list) {
			/*
			 * Here unmap happens in parallel for only one reason:
			 * blk_cleanup_queue() takes around half a second, so
			 * on huge amount of devices the whole module unload
			 * procedure takes minutes.
			 */
			INIT_WORK(&dev->unmap_on_rmmod_work, unmap_device_work);
			queue_work(system_long_wq, &dev->unmap_on_rmmod_work);
		}
		rnbd_clt_put_sess(sess);
	}
	/* Wait for all scheduled unmap works */
	flush_workqueue(system_long_wq);
	WARN_ON(!list_empty(&sess_list));
}

static int __init rnbd_client_init(void)
{
	int err = 0;

	BUILD_BUG_ON(sizeof(struct rnbd_msg_hdr) != 4);
	BUILD_BUG_ON(sizeof(struct rnbd_msg_sess_info) != 36);
	BUILD_BUG_ON(sizeof(struct rnbd_msg_sess_info_rsp) != 36);
	BUILD_BUG_ON(sizeof(struct rnbd_msg_open) != 264);
	BUILD_BUG_ON(sizeof(struct rnbd_msg_close) != 8);
	BUILD_BUG_ON(sizeof(struct rnbd_msg_open_rsp) != 56);
	rnbd_client_major = register_blkdev(rnbd_client_major, "rnbd");
	if (rnbd_client_major <= 0) {
		pr_err("Failed to load module, block device registration failed\n");
		return -EBUSY;
	}

	err = rnbd_clt_create_sysfs_files();
	if (err) {
		pr_err("Failed to load module, creating sysfs device files failed, err: %d\n",
		       err);
		unregister_blkdev(rnbd_client_major, "rnbd");
	}

	return err;
}

static void __exit rnbd_client_exit(void)
{
	rnbd_destroy_sessions();
	unregister_blkdev(rnbd_client_major, "rnbd");
	ida_destroy(&index_ida);
}

module_init(rnbd_client_init);
module_exit(rnbd_client_exit);