summaryrefslogtreecommitdiffstats
path: root/module/zfs/zvol.c
blob: 79c56cd78188f2cdb31f182d4725136f97ea4b79 (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
/*
 * 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 (C) 2008-2010 Lawrence Livermore National Security, LLC.
 * Produced at Lawrence Livermore National Laboratory (cf, DISCLAIMER).
 * Rewritten for Linux by Brian Behlendorf <behlendorf1@llnl.gov>.
 * LLNL-CODE-403049.
 *
 * ZFS volume emulation driver.
 *
 * Makes a DMU object look like a volume of arbitrary size, up to 2^64 bytes.
 * Volumes are accessed through the symbolic links named:
 *
 * /dev/<pool_name>/<dataset_name>
 *
 * Volumes are persistent through reboot and module load.  No user command
 * needs to be run before opening and using a device.
 */

#include <sys/dbuf.h>
#include <sys/dmu_traverse.h>
#include <sys/dsl_dataset.h>
#include <sys/dsl_prop.h>
#include <sys/zap.h>
#include <sys/zil_impl.h>
#include <sys/zio.h>
#include <sys/zfs_rlock.h>
#include <sys/zfs_znode.h>
#include <sys/zvol.h>
#include <linux/blkdev_compat.h>

unsigned int zvol_inhibit_dev = 0;
unsigned int zvol_major = ZVOL_MAJOR;
unsigned int zvol_threads = 32;
unsigned long zvol_max_discard_blocks = 16384;

static taskq_t *zvol_taskq;
static kmutex_t zvol_state_lock;
static list_t zvol_state_list;
static char *zvol_tag = "zvol_tag";

/*
 * The in-core state of each volume.
 */
typedef struct zvol_state {
	char			zv_name[MAXNAMELEN];	/* name */
	uint64_t		zv_volsize;	/* advertised space */
	uint64_t		zv_volblocksize;/* volume block size */
	objset_t		*zv_objset;	/* objset handle */
	uint32_t		zv_flags;	/* ZVOL_* flags */
	uint32_t		zv_open_count;	/* open counts */
	uint32_t		zv_changed;	/* disk changed */
	zilog_t			*zv_zilog;	/* ZIL handle */
	znode_t			zv_znode;	/* for range locking */
	dmu_buf_t		*zv_dbuf;	/* bonus handle */
	dev_t			zv_dev;		/* device id */
	struct gendisk		*zv_disk;	/* generic disk */
	struct request_queue	*zv_queue;	/* request queue */
	spinlock_t		zv_lock;	/* request queue lock */
	list_node_t		zv_next;	/* next zvol_state_t linkage */
} zvol_state_t;

#define	ZVOL_RDONLY	0x1

/*
 * Find the next available range of ZVOL_MINORS minor numbers.  The
 * zvol_state_list is kept in ascending minor order so we simply need
 * to scan the list for the first gap in the sequence.  This allows us
 * to recycle minor number as devices are created and removed.
 */
static int
zvol_find_minor(unsigned *minor)
{
	zvol_state_t *zv;

	*minor = 0;
	ASSERT(MUTEX_HELD(&zvol_state_lock));
	for (zv = list_head(&zvol_state_list); zv != NULL;
	     zv = list_next(&zvol_state_list, zv), *minor += ZVOL_MINORS) {
		if (MINOR(zv->zv_dev) != MINOR(*minor))
			break;
	}

	/* All minors are in use */
	if (*minor >= (1 << MINORBITS))
		return ENXIO;

	return 0;
}

/*
 * Find a zvol_state_t given the full major+minor dev_t.
 */
static zvol_state_t *
zvol_find_by_dev(dev_t dev)
{
	zvol_state_t *zv;

	ASSERT(MUTEX_HELD(&zvol_state_lock));
	for (zv = list_head(&zvol_state_list); zv != NULL;
	     zv = list_next(&zvol_state_list, zv)) {
		if (zv->zv_dev == dev)
			return zv;
	}

	return NULL;
}

/*
 * Find a zvol_state_t given the name provided at zvol_alloc() time.
 */
static zvol_state_t *
zvol_find_by_name(const char *name)
{
	zvol_state_t *zv;

	ASSERT(MUTEX_HELD(&zvol_state_lock));
	for (zv = list_head(&zvol_state_list); zv != NULL;
	     zv = list_next(&zvol_state_list, zv)) {
		if (!strncmp(zv->zv_name, name, MAXNAMELEN))
			return zv;
	}

	return NULL;
}


/*
 * Given a path, return TRUE if path is a ZVOL.
 */
boolean_t
zvol_is_zvol(const char *device)
{
	struct block_device *bdev;
	unsigned int major;

	bdev = lookup_bdev(device);
	if (IS_ERR(bdev))
		return (B_FALSE);

	major = MAJOR(bdev->bd_dev);
	bdput(bdev);

	if (major == zvol_major)
            return (B_TRUE);

	return (B_FALSE);
}

/*
 * ZFS_IOC_CREATE callback handles dmu zvol and zap object creation.
 */
void
zvol_create_cb(objset_t *os, void *arg, cred_t *cr, dmu_tx_t *tx)
{
	zfs_creat_t *zct = arg;
	nvlist_t *nvprops = zct->zct_props;
	int error;
	uint64_t volblocksize, volsize;

	VERIFY(nvlist_lookup_uint64(nvprops,
	    zfs_prop_to_name(ZFS_PROP_VOLSIZE), &volsize) == 0);
	if (nvlist_lookup_uint64(nvprops,
	    zfs_prop_to_name(ZFS_PROP_VOLBLOCKSIZE), &volblocksize) != 0)
		volblocksize = zfs_prop_default_numeric(ZFS_PROP_VOLBLOCKSIZE);

	/*
	 * These properties must be removed from the list so the generic
	 * property setting step won't apply to them.
	 */
	VERIFY(nvlist_remove_all(nvprops,
	    zfs_prop_to_name(ZFS_PROP_VOLSIZE)) == 0);
	(void) nvlist_remove_all(nvprops,
	    zfs_prop_to_name(ZFS_PROP_VOLBLOCKSIZE));

	error = dmu_object_claim(os, ZVOL_OBJ, DMU_OT_ZVOL, volblocksize,
	    DMU_OT_NONE, 0, tx);
	ASSERT(error == 0);

	error = zap_create_claim(os, ZVOL_ZAP_OBJ, DMU_OT_ZVOL_PROP,
	    DMU_OT_NONE, 0, tx);
	ASSERT(error == 0);

	error = zap_update(os, ZVOL_ZAP_OBJ, "size", 8, 1, &volsize, tx);
	ASSERT(error == 0);
}

/*
 * ZFS_IOC_OBJSET_STATS entry point.
 */
int
zvol_get_stats(objset_t *os, nvlist_t *nv)
{
	int error;
	dmu_object_info_t *doi;
	uint64_t val;

	error = zap_lookup(os, ZVOL_ZAP_OBJ, "size", 8, 1, &val);
	if (error)
		return (error);

	dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_VOLSIZE, val);
	doi = kmem_alloc(sizeof(dmu_object_info_t), KM_SLEEP);
	error = dmu_object_info(os, ZVOL_OBJ, doi);

	if (error == 0) {
		dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_VOLBLOCKSIZE,
		    doi->doi_data_block_size);
	}

	kmem_free(doi, sizeof(dmu_object_info_t));

	return (error);
}

/*
 * Sanity check volume size.
 */
int
zvol_check_volsize(uint64_t volsize, uint64_t blocksize)
{
	if (volsize == 0)
		return (SET_ERROR(EINVAL));

	if (volsize % blocksize != 0)
		return (SET_ERROR(EINVAL));

#ifdef _ILP32
	if (volsize - 1 > MAXOFFSET_T)
		return (SET_ERROR(EOVERFLOW));
#endif
	return (0);
}

/*
 * Ensure the zap is flushed then inform the VFS of the capacity change.
 */
static int
zvol_update_volsize(zvol_state_t *zv, uint64_t volsize, objset_t *os)
{
	struct block_device *bdev;
	dmu_tx_t *tx;
	int error;

	ASSERT(MUTEX_HELD(&zvol_state_lock));

	tx = dmu_tx_create(os);
	dmu_tx_hold_zap(tx, ZVOL_ZAP_OBJ, TRUE, NULL);
	error = dmu_tx_assign(tx, TXG_WAIT);
	if (error) {
		dmu_tx_abort(tx);
		return (error);
	}

	error = zap_update(os, ZVOL_ZAP_OBJ, "size", 8, 1,
	    &volsize, tx);
	dmu_tx_commit(tx);

	if (error)
		return (error);

	error = dmu_free_long_range(os,
	    ZVOL_OBJ, volsize, DMU_OBJECT_END);
	if (error)
		return (error);

	bdev = bdget_disk(zv->zv_disk, 0);
	if (!bdev)
		return (SET_ERROR(EIO));
/*
 * 2.6.28 API change
 * Added check_disk_size_change() helper function.
 */
#ifdef HAVE_CHECK_DISK_SIZE_CHANGE
	set_capacity(zv->zv_disk, volsize >> 9);
	zv->zv_volsize = volsize;
	check_disk_size_change(zv->zv_disk, bdev);
#else
	zv->zv_volsize = volsize;
	zv->zv_changed = 1;
	(void) check_disk_change(bdev);
#endif /* HAVE_CHECK_DISK_SIZE_CHANGE */

	bdput(bdev);

	return (0);
}

/*
 * Set ZFS_PROP_VOLSIZE set entry point.
 */
int
zvol_set_volsize(const char *name, uint64_t volsize)
{
	zvol_state_t *zv;
	dmu_object_info_t *doi;
	objset_t *os = NULL;
	uint64_t readonly;
	int error;

	error = dsl_prop_get_integer(name,
	    zfs_prop_to_name(ZFS_PROP_READONLY), &readonly, NULL);
	if (error != 0)
		return (error);
	if (readonly)
		return (SET_ERROR(EROFS));

	mutex_enter(&zvol_state_lock);

	zv = zvol_find_by_name(name);
	if (zv == NULL) {
		error = SET_ERROR(ENXIO);
		goto out;
	}

	doi = kmem_alloc(sizeof(dmu_object_info_t), KM_SLEEP);

	error = dmu_objset_hold(name, FTAG, &os);
	if (error)
		goto out_doi;

	if ((error = dmu_object_info(os, ZVOL_OBJ, doi)) != 0 ||
	    (error = zvol_check_volsize(volsize,doi->doi_data_block_size)) != 0)
		goto out_doi;

	VERIFY(dsl_prop_get_integer(name, "readonly", &readonly, NULL) == 0);
	if (readonly) {
		error = SET_ERROR(EROFS);
		goto out_doi;
	}

	if (get_disk_ro(zv->zv_disk) || (zv->zv_flags & ZVOL_RDONLY)) {
		error = SET_ERROR(EROFS);
		goto out_doi;
	}

	error = zvol_update_volsize(zv, volsize, os);
out_doi:
	kmem_free(doi, sizeof(dmu_object_info_t));
out:
	if (os)
		dmu_objset_rele(os, FTAG);

	mutex_exit(&zvol_state_lock);

	return (error);
}

/*
 * Sanity check volume block size.
 */
int
zvol_check_volblocksize(uint64_t volblocksize)
{
	if (volblocksize < SPA_MINBLOCKSIZE ||
	    volblocksize > SPA_MAXBLOCKSIZE ||
	    !ISP2(volblocksize))
		return (SET_ERROR(EDOM));

	return (0);
}

/*
 * Set ZFS_PROP_VOLBLOCKSIZE set entry point.
 */
int
zvol_set_volblocksize(const char *name, uint64_t volblocksize)
{
	zvol_state_t *zv;
	dmu_tx_t *tx;
	int error;

	mutex_enter(&zvol_state_lock);

	zv = zvol_find_by_name(name);
	if (zv == NULL) {
		error = SET_ERROR(ENXIO);
		goto out;
	}

	if (get_disk_ro(zv->zv_disk) || (zv->zv_flags & ZVOL_RDONLY)) {
		error = SET_ERROR(EROFS);
		goto out;
	}

	tx = dmu_tx_create(zv->zv_objset);
	dmu_tx_hold_bonus(tx, ZVOL_OBJ);
	error = dmu_tx_assign(tx, TXG_WAIT);
	if (error) {
		dmu_tx_abort(tx);
	} else {
		error = dmu_object_set_blocksize(zv->zv_objset, ZVOL_OBJ,
		    volblocksize, 0, tx);
		if (error == ENOTSUP)
			error = SET_ERROR(EBUSY);
		dmu_tx_commit(tx);
		if (error == 0)
			zv->zv_volblocksize = volblocksize;
	}
out:
	mutex_exit(&zvol_state_lock);

	return (error);
}

/*
 * Replay a TX_WRITE ZIL transaction that didn't get committed
 * after a system failure
 */
static int
zvol_replay_write(zvol_state_t *zv, lr_write_t *lr, boolean_t byteswap)
{
	objset_t *os = zv->zv_objset;
	char *data = (char *)(lr + 1);	/* data follows lr_write_t */
	uint64_t off = lr->lr_offset;
	uint64_t len = lr->lr_length;
	dmu_tx_t *tx;
	int error;

	if (byteswap)
		byteswap_uint64_array(lr, sizeof (*lr));

	tx = dmu_tx_create(os);
	dmu_tx_hold_write(tx, ZVOL_OBJ, off, len);
	error = dmu_tx_assign(tx, TXG_WAIT);
	if (error) {
		dmu_tx_abort(tx);
	} else {
		dmu_write(os, ZVOL_OBJ, off, len, data, tx);
		dmu_tx_commit(tx);
	}

	return (error);
}

static int
zvol_replay_err(zvol_state_t *zv, lr_t *lr, boolean_t byteswap)
{
	return (SET_ERROR(ENOTSUP));
}

/*
 * Callback vectors for replaying records.
 * Only TX_WRITE is needed for zvol.
 */
zil_replay_func_t zvol_replay_vector[TX_MAX_TYPE] = {
	(zil_replay_func_t)zvol_replay_err,	/* no such transaction type */
	(zil_replay_func_t)zvol_replay_err,	/* TX_CREATE */
	(zil_replay_func_t)zvol_replay_err,	/* TX_MKDIR */
	(zil_replay_func_t)zvol_replay_err,	/* TX_MKXATTR */
	(zil_replay_func_t)zvol_replay_err,	/* TX_SYMLINK */
	(zil_replay_func_t)zvol_replay_err,	/* TX_REMOVE */
	(zil_replay_func_t)zvol_replay_err,	/* TX_RMDIR */
	(zil_replay_func_t)zvol_replay_err,	/* TX_LINK */
	(zil_replay_func_t)zvol_replay_err,	/* TX_RENAME */
	(zil_replay_func_t)zvol_replay_write,	/* TX_WRITE */
	(zil_replay_func_t)zvol_replay_err,	/* TX_TRUNCATE */
	(zil_replay_func_t)zvol_replay_err,	/* TX_SETATTR */
	(zil_replay_func_t)zvol_replay_err,	/* TX_ACL */
};

/*
 * zvol_log_write() handles synchronous writes using TX_WRITE ZIL transactions.
 *
 * We store data in the log buffers if it's small enough.
 * Otherwise we will later flush the data out via dmu_sync().
 */
ssize_t zvol_immediate_write_sz = 32768;

static void
zvol_log_write(zvol_state_t *zv, dmu_tx_t *tx,
	       uint64_t offset, uint64_t size, int sync)
{
	uint32_t blocksize = zv->zv_volblocksize;
	zilog_t *zilog = zv->zv_zilog;
	boolean_t slogging;
	ssize_t immediate_write_sz;

	if (zil_replaying(zilog, tx))
		return;

	immediate_write_sz = (zilog->zl_logbias == ZFS_LOGBIAS_THROUGHPUT)
		? 0 : zvol_immediate_write_sz;
	slogging = spa_has_slogs(zilog->zl_spa) &&
		(zilog->zl_logbias == ZFS_LOGBIAS_LATENCY);

	while (size) {
		itx_t *itx;
		lr_write_t *lr;
		ssize_t len;
		itx_wr_state_t write_state;

		/*
		 * Unlike zfs_log_write() we can be called with
		 * up to DMU_MAX_ACCESS/2 (5MB) writes.
		 */
		if (blocksize > immediate_write_sz && !slogging &&
		    size >= blocksize && offset % blocksize == 0) {
			write_state = WR_INDIRECT; /* uses dmu_sync */
			len = blocksize;
		} else if (sync) {
			write_state = WR_COPIED;
			len = MIN(ZIL_MAX_LOG_DATA, size);
		} else {
			write_state = WR_NEED_COPY;
			len = MIN(ZIL_MAX_LOG_DATA, size);
		}

		itx = zil_itx_create(TX_WRITE, sizeof (*lr) +
		    (write_state == WR_COPIED ? len : 0));
		lr = (lr_write_t *)&itx->itx_lr;
		if (write_state == WR_COPIED && dmu_read(zv->zv_objset,
		    ZVOL_OBJ, offset, len, lr+1, DMU_READ_NO_PREFETCH) != 0) {
			zil_itx_destroy(itx);
			itx = zil_itx_create(TX_WRITE, sizeof (*lr));
			lr = (lr_write_t *)&itx->itx_lr;
			write_state = WR_NEED_COPY;
		}

		itx->itx_wr_state = write_state;
		if (write_state == WR_NEED_COPY)
			itx->itx_sod += len;
		lr->lr_foid = ZVOL_OBJ;
		lr->lr_offset = offset;
		lr->lr_length = len;
		lr->lr_blkoff = 0;
		BP_ZERO(&lr->lr_blkptr);

		itx->itx_private = zv;
		itx->itx_sync = sync;

		(void) zil_itx_assign(zilog, itx, tx);

		offset += len;
		size -= len;
	}
}

/*
 * Common write path running under the zvol taskq context.  This function
 * is responsible for copying the request structure data in to the DMU and
 * signaling the request queue with the result of the copy.
 */
static void
zvol_write(void *arg)
{
	struct request *req = (struct request *)arg;
	struct request_queue *q = req->q;
	zvol_state_t *zv = q->queuedata;
	uint64_t offset = blk_rq_pos(req) << 9;
	uint64_t size = blk_rq_bytes(req);
	int error = 0;
	dmu_tx_t *tx;
	rl_t *rl;

	/*
	 * Annotate this call path with a flag that indicates that it is
	 * unsafe to use KM_SLEEP during memory allocations due to the
	 * potential for a deadlock.  KM_PUSHPAGE should be used instead.
	 */
	ASSERT(!(current->flags & PF_NOFS));
	current->flags |= PF_NOFS;

	if (req->cmd_flags & VDEV_REQ_FLUSH)
		zil_commit(zv->zv_zilog, ZVOL_OBJ);

	/*
	 * Some requests are just for flush and nothing else.
	 */
	if (size == 0) {
		blk_end_request(req, 0, size);
		goto out;
	}

	rl = zfs_range_lock(&zv->zv_znode, offset, size, RL_WRITER);

	tx = dmu_tx_create(zv->zv_objset);
	dmu_tx_hold_write(tx, ZVOL_OBJ, offset, size);

	/* This will only fail for ENOSPC */
	error = dmu_tx_assign(tx, TXG_WAIT);
	if (error) {
		dmu_tx_abort(tx);
		zfs_range_unlock(rl);
		blk_end_request(req, -error, size);
		goto out;
	}

	error = dmu_write_req(zv->zv_objset, ZVOL_OBJ, req, tx);
	if (error == 0)
		zvol_log_write(zv, tx, offset, size,
		    req->cmd_flags & VDEV_REQ_FUA);

	dmu_tx_commit(tx);
	zfs_range_unlock(rl);

	if ((req->cmd_flags & VDEV_REQ_FUA) ||
	    zv->zv_objset->os_sync == ZFS_SYNC_ALWAYS)
		zil_commit(zv->zv_zilog, ZVOL_OBJ);

	blk_end_request(req, -error, size);
out:
	current->flags &= ~PF_NOFS;
}

#ifdef HAVE_BLK_QUEUE_DISCARD
static void
zvol_discard(void *arg)
{
	struct request *req = (struct request *)arg;
	struct request_queue *q = req->q;
	zvol_state_t *zv = q->queuedata;
	uint64_t start = blk_rq_pos(req) << 9;
	uint64_t end = start + blk_rq_bytes(req);
	int error;
	rl_t *rl;

	/*
	 * Annotate this call path with a flag that indicates that it is
	 * unsafe to use KM_SLEEP during memory allocations due to the
	 * potential for a deadlock.  KM_PUSHPAGE should be used instead.
	 */
	ASSERT(!(current->flags & PF_NOFS));
	current->flags |= PF_NOFS;

	if (end > zv->zv_volsize) {
		blk_end_request(req, -EIO, blk_rq_bytes(req));
		goto out;
	}

	/*
	 * Align the request to volume block boundaries. If we don't,
	 * then this will force dnode_free_range() to zero out the
	 * unaligned parts, which is slow (read-modify-write) and
	 * useless since we are not freeing any space by doing so.
	 */
	start = P2ROUNDUP(start, zv->zv_volblocksize);
	end = P2ALIGN(end, zv->zv_volblocksize);

	if (start >= end) {
		blk_end_request(req, 0, blk_rq_bytes(req));
		goto out;
	}

	rl = zfs_range_lock(&zv->zv_znode, start, end - start, RL_WRITER);

	error = dmu_free_long_range(zv->zv_objset, ZVOL_OBJ, start, end - start);

	/*
	 * TODO: maybe we should add the operation to the log.
	 */

	zfs_range_unlock(rl);

	blk_end_request(req, -error, blk_rq_bytes(req));
out:
	current->flags &= ~PF_NOFS;
}
#endif /* HAVE_BLK_QUEUE_DISCARD */

/*
 * Common read path running under the zvol taskq context.  This function
 * is responsible for copying the requested data out of the DMU and in to
 * a linux request structure.  It then must signal the request queue with
 * an error code describing the result of the copy.
 */
static void
zvol_read(void *arg)
{
	struct request *req = (struct request *)arg;
	struct request_queue *q = req->q;
	zvol_state_t *zv = q->queuedata;
	uint64_t offset = blk_rq_pos(req) << 9;
	uint64_t size = blk_rq_bytes(req);
	int error;
	rl_t *rl;

	if (size == 0) {
		blk_end_request(req, 0, size);
		return;
	}

	rl = zfs_range_lock(&zv->zv_znode, offset, size, RL_READER);

	error = dmu_read_req(zv->zv_objset, ZVOL_OBJ, req);

	zfs_range_unlock(rl);

	/* convert checksum errors into IO errors */
	if (error == ECKSUM)
		error = SET_ERROR(EIO);

	blk_end_request(req, -error, size);
}

/*
 * Request will be added back to the request queue and retried if
 * it cannot be immediately dispatched to the taskq for handling
 */
static inline void
zvol_dispatch(task_func_t func, struct request *req)
{
	if (!taskq_dispatch(zvol_taskq, func, (void *)req, TQ_NOSLEEP))
		blk_requeue_request(req->q, req);
}

/*
 * Common request path.  Rather than registering a custom make_request()
 * function we use the generic Linux version.  This is done because it allows
 * us to easily merge read requests which would otherwise we performed
 * synchronously by the DMU.  This is less critical in write case where the
 * DMU will perform the correct merging within a transaction group.  Using
 * the generic make_request() also let's use leverage the fact that the
 * elevator with ensure correct ordering in regards to barrior IOs.  On
 * the downside it means that in the write case we end up doing request
 * merging twice once in the elevator and once in the DMU.
 *
 * The request handler is called under a spin lock so all the real work
 * is handed off to be done in the context of the zvol taskq.  This function
 * simply performs basic request sanity checking and hands off the request.
 */
static void
zvol_request(struct request_queue *q)
{
	zvol_state_t *zv = q->queuedata;
	struct request *req;
	unsigned int size;

	while ((req = blk_fetch_request(q)) != NULL) {
		size = blk_rq_bytes(req);

		if (size != 0 && blk_rq_pos(req) + blk_rq_sectors(req) >
		    get_capacity(zv->zv_disk)) {
			printk(KERN_INFO
			       "%s: bad access: block=%llu, count=%lu\n",
			       req->rq_disk->disk_name,
			       (long long unsigned)blk_rq_pos(req),
			       (long unsigned)blk_rq_sectors(req));
			__blk_end_request(req, -EIO, size);
			continue;
		}

		if (!blk_fs_request(req)) {
			printk(KERN_INFO "%s: non-fs cmd\n",
			       req->rq_disk->disk_name);
			__blk_end_request(req, -EIO, size);
			continue;
		}

		switch (rq_data_dir(req)) {
		case READ:
			zvol_dispatch(zvol_read, req);
			break;
		case WRITE:
			if (unlikely(get_disk_ro(zv->zv_disk)) ||
			    unlikely(zv->zv_flags & ZVOL_RDONLY)) {
				__blk_end_request(req, -EROFS, size);
				break;
			}

#ifdef HAVE_BLK_QUEUE_DISCARD
			if (req->cmd_flags & VDEV_REQ_DISCARD) {
				zvol_dispatch(zvol_discard, req);
				break;
			}
#endif /* HAVE_BLK_QUEUE_DISCARD */

			zvol_dispatch(zvol_write, req);
			break;
		default:
			printk(KERN_INFO "%s: unknown cmd: %d\n",
			       req->rq_disk->disk_name, (int)rq_data_dir(req));
			__blk_end_request(req, -EIO, size);
			break;
		}
	}
}

static void
zvol_get_done(zgd_t *zgd, int error)
{
	if (zgd->zgd_db)
		dmu_buf_rele(zgd->zgd_db, zgd);

	zfs_range_unlock(zgd->zgd_rl);

	if (error == 0 && zgd->zgd_bp)
		zil_add_block(zgd->zgd_zilog, zgd->zgd_bp);

	kmem_free(zgd, sizeof (zgd_t));
}

/*
 * Get data to generate a TX_WRITE intent log record.
 */
static int
zvol_get_data(void *arg, lr_write_t *lr, char *buf, zio_t *zio)
{
	zvol_state_t *zv = arg;
	objset_t *os = zv->zv_objset;
	uint64_t object = ZVOL_OBJ;
	uint64_t offset = lr->lr_offset;
	uint64_t size = lr->lr_length;
	blkptr_t *bp = &lr->lr_blkptr;
	dmu_buf_t *db;
	zgd_t *zgd;
	int error;

	ASSERT(zio != NULL);
	ASSERT(size != 0);

	zgd = (zgd_t *)kmem_zalloc(sizeof (zgd_t), KM_PUSHPAGE);
	zgd->zgd_zilog = zv->zv_zilog;
	zgd->zgd_rl = zfs_range_lock(&zv->zv_znode, offset, size, RL_READER);

	/*
	 * Write records come in two flavors: immediate and indirect.
	 * For small writes it's cheaper to store the data with the
	 * log record (immediate); for large writes it's cheaper to
	 * sync the data and get a pointer to it (indirect) so that
	 * we don't have to write the data twice.
	 */
	if (buf != NULL) { /* immediate write */
		error = dmu_read(os, object, offset, size, buf,
		    DMU_READ_NO_PREFETCH);
	} else {
		size = zv->zv_volblocksize;
		offset = P2ALIGN_TYPED(offset, size, uint64_t);
		error = dmu_buf_hold(os, object, offset, zgd, &db,
		    DMU_READ_NO_PREFETCH);
		if (error == 0) {
			blkptr_t *obp = dmu_buf_get_blkptr(db);
			if (obp) {
				ASSERT(BP_IS_HOLE(bp));
				*bp = *obp;
			}

			zgd->zgd_db = db;
			zgd->zgd_bp = &lr->lr_blkptr;

			ASSERT(db != NULL);
			ASSERT(db->db_offset == offset);
			ASSERT(db->db_size == size);

			error = dmu_sync(zio, lr->lr_common.lrc_txg,
			    zvol_get_done, zgd);

			if (error == 0)
				return (0);
		}
	}

	zvol_get_done(zgd, error);

	return (error);
}

/*
 * The zvol_state_t's are inserted in increasing MINOR(dev_t) order.
 */
static void
zvol_insert(zvol_state_t *zv_insert)
{
	zvol_state_t *zv = NULL;

	ASSERT(MUTEX_HELD(&zvol_state_lock));
	ASSERT3U(MINOR(zv_insert->zv_dev) & ZVOL_MINOR_MASK, ==, 0);
	for (zv = list_head(&zvol_state_list); zv != NULL;
	     zv = list_next(&zvol_state_list, zv)) {
		if (MINOR(zv->zv_dev) > MINOR(zv_insert->zv_dev))
			break;
	}

	list_insert_before(&zvol_state_list, zv, zv_insert);
}

/*
 * Simply remove the zvol from to list of zvols.
 */
static void
zvol_remove(zvol_state_t *zv_remove)
{
	ASSERT(MUTEX_HELD(&zvol_state_lock));
	list_remove(&zvol_state_list, zv_remove);
}

static int
zvol_first_open(zvol_state_t *zv)
{
	objset_t *os;
	uint64_t volsize;
	int locked = 0;
	int error;
	uint64_t ro;

	/*
	 * In all other cases the spa_namespace_lock is taken before the
	 * bdev->bd_mutex lock.  But in this case the Linux __blkdev_get()
	 * function calls fops->open() with the bdev->bd_mutex lock held.
	 *
	 * To avoid a potential lock inversion deadlock we preemptively
	 * try to take the spa_namespace_lock().  Normally it will not
	 * be contended and this is safe because spa_open_common() handles
	 * the case where the caller already holds the spa_namespace_lock.
	 *
	 * When it is contended we risk a lock inversion if we were to
	 * block waiting for the lock.  Luckily, the __blkdev_get()
	 * function allows us to return -ERESTARTSYS which will result in
	 * bdev->bd_mutex being dropped, reacquired, and fops->open() being
	 * called again.  This process can be repeated safely until both
	 * locks are acquired.
	 */
	if (!mutex_owned(&spa_namespace_lock)) {
		locked = mutex_tryenter(&spa_namespace_lock);
		if (!locked)
			return (-SET_ERROR(ERESTARTSYS));
	}

	/* lie and say we're read-only */
	error = dmu_objset_own(zv->zv_name, DMU_OST_ZVOL, 1, zvol_tag, &os);
	if (error)
		goto out_mutex;

	error = zap_lookup(os, ZVOL_ZAP_OBJ, "size", 8, 1, &volsize);
	if (error) {
		dmu_objset_disown(os, zvol_tag);
		goto out_mutex;
	}

	zv->zv_objset = os;
	error = dmu_bonus_hold(os, ZVOL_OBJ, zvol_tag, &zv->zv_dbuf);
	if (error) {
		dmu_objset_disown(os, zvol_tag);
		goto out_mutex;
	}

	set_capacity(zv->zv_disk, volsize >> 9);
	zv->zv_volsize = volsize;
	zv->zv_zilog = zil_open(os, zvol_get_data);

	VERIFY(dsl_prop_get_integer(zv->zv_name, "readonly", &ro, NULL) == 0);
	if (ro || dmu_objset_is_snapshot(os) ||
	    !spa_writeable(dmu_objset_spa(os))) {
		set_disk_ro(zv->zv_disk, 1);
		zv->zv_flags |= ZVOL_RDONLY;
	} else {
		set_disk_ro(zv->zv_disk, 0);
		zv->zv_flags &= ~ZVOL_RDONLY;
	}

out_mutex:
	if (locked)
		mutex_exit(&spa_namespace_lock);

	return (-error);
}

static void
zvol_last_close(zvol_state_t *zv)
{
	zil_close(zv->zv_zilog);
	zv->zv_zilog = NULL;

	dmu_buf_rele(zv->zv_dbuf, zvol_tag);
	zv->zv_dbuf = NULL;

	/*
	 * Evict cached data
	 */
	if (dsl_dataset_is_dirty(dmu_objset_ds(zv->zv_objset)) &&
	    !(zv->zv_flags & ZVOL_RDONLY))
		txg_wait_synced(dmu_objset_pool(zv->zv_objset), 0);
	(void) dmu_objset_evict_dbufs(zv->zv_objset);

	dmu_objset_disown(zv->zv_objset, zvol_tag);
	zv->zv_objset = NULL;
}

static int
zvol_open(struct block_device *bdev, fmode_t flag)
{
	zvol_state_t *zv = bdev->bd_disk->private_data;
	int error = 0, drop_mutex = 0;

	/*
	 * If the caller is already holding the mutex do not take it
	 * again, this will happen as part of zvol_create_minor().
	 * Once add_disk() is called the device is live and the kernel
	 * will attempt to open it to read the partition information.
	 */
	if (!mutex_owned(&zvol_state_lock)) {
		mutex_enter(&zvol_state_lock);
		drop_mutex = 1;
	}

	ASSERT3P(zv, !=, NULL);

	if (zv->zv_open_count == 0) {
		error = zvol_first_open(zv);
		if (error)
			goto out_mutex;
	}

	if ((flag & FMODE_WRITE) &&
	    (get_disk_ro(zv->zv_disk) || (zv->zv_flags & ZVOL_RDONLY))) {
		error = -EROFS;
		goto out_open_count;
	}

	zv->zv_open_count++;

out_open_count:
	if (zv->zv_open_count == 0)
		zvol_last_close(zv);

out_mutex:
	if (drop_mutex)
		mutex_exit(&zvol_state_lock);

	check_disk_change(bdev);

	return (error);
}

#ifdef HAVE_BLOCK_DEVICE_OPERATIONS_RELEASE_VOID
static void
#else
static int
#endif
zvol_release(struct gendisk *disk, fmode_t mode)
{
	zvol_state_t *zv = disk->private_data;
	int drop_mutex = 0;

	if (!mutex_owned(&zvol_state_lock)) {
		mutex_enter(&zvol_state_lock);
		drop_mutex = 1;
	}

	ASSERT3P(zv, !=, NULL);
	ASSERT3U(zv->zv_open_count, >, 0);
	zv->zv_open_count--;
	if (zv->zv_open_count == 0)
		zvol_last_close(zv);

	if (drop_mutex)
		mutex_exit(&zvol_state_lock);

#ifndef HAVE_BLOCK_DEVICE_OPERATIONS_RELEASE_VOID
	return (0);
#endif
}

static int
zvol_ioctl(struct block_device *bdev, fmode_t mode,
           unsigned int cmd, unsigned long arg)
{
	zvol_state_t *zv = bdev->bd_disk->private_data;
	int error = 0;

	if (zv == NULL)
		return (-SET_ERROR(ENXIO));

	switch (cmd) {
	case BLKFLSBUF:
		zil_commit(zv->zv_zilog, ZVOL_OBJ);
		break;
	case BLKZNAME:
		error = copy_to_user((void *)arg, zv->zv_name, MAXNAMELEN);
		break;

	default:
		error = -ENOTTY;
		break;

	}

	return (error);
}

#ifdef CONFIG_COMPAT
static int
zvol_compat_ioctl(struct block_device *bdev, fmode_t mode,
                  unsigned cmd, unsigned long arg)
{
	return zvol_ioctl(bdev, mode, cmd, arg);
}
#else
#define zvol_compat_ioctl   NULL
#endif

static int zvol_media_changed(struct gendisk *disk)
{
	zvol_state_t *zv = disk->private_data;

	return zv->zv_changed;
}

static int zvol_revalidate_disk(struct gendisk *disk)
{
	zvol_state_t *zv = disk->private_data;

	zv->zv_changed = 0;
	set_capacity(zv->zv_disk, zv->zv_volsize >> 9);

	return 0;
}

/*
 * Provide a simple virtual geometry for legacy compatibility.  For devices
 * smaller than 1 MiB a small head and sector count is used to allow very
 * tiny devices.  For devices over 1 Mib a standard head and sector count
 * is used to keep the cylinders count reasonable.
 */
static int
zvol_getgeo(struct block_device *bdev, struct hd_geometry *geo)
{
	zvol_state_t *zv = bdev->bd_disk->private_data;
	sector_t sectors = get_capacity(zv->zv_disk);

	if (sectors > 2048) {
		geo->heads = 16;
		geo->sectors = 63;
	} else {
		geo->heads = 2;
		geo->sectors = 4;
	}

	geo->start = 0;
	geo->cylinders = sectors / (geo->heads * geo->sectors);

	return 0;
}

static struct kobject *
zvol_probe(dev_t dev, int *part, void *arg)
{
	zvol_state_t *zv;
	struct kobject *kobj;

	mutex_enter(&zvol_state_lock);
	zv = zvol_find_by_dev(dev);
	kobj = zv ? get_disk(zv->zv_disk) : NULL;
	mutex_exit(&zvol_state_lock);

	return kobj;
}

#ifdef HAVE_BDEV_BLOCK_DEVICE_OPERATIONS
static struct block_device_operations zvol_ops = {
	.open            = zvol_open,
	.release         = zvol_release,
	.ioctl           = zvol_ioctl,
	.compat_ioctl    = zvol_compat_ioctl,
	.media_changed   = zvol_media_changed,
	.revalidate_disk = zvol_revalidate_disk,
	.getgeo          = zvol_getgeo,
        .owner           = THIS_MODULE,
};

#else /* HAVE_BDEV_BLOCK_DEVICE_OPERATIONS */

static int
zvol_open_by_inode(struct inode *inode, struct file *file)
{
	return zvol_open(inode->i_bdev, file->f_mode);
}

static int
zvol_release_by_inode(struct inode *inode, struct file *file)
{
	return zvol_release(inode->i_bdev->bd_disk, file->f_mode);
}

static int
zvol_ioctl_by_inode(struct inode *inode, struct file *file,
                    unsigned int cmd, unsigned long arg)
{
	if (file == NULL || inode == NULL)
		return -EINVAL;
	return zvol_ioctl(inode->i_bdev, file->f_mode, cmd, arg);
}

# ifdef CONFIG_COMPAT
static long
zvol_compat_ioctl_by_inode(struct file *file,
                           unsigned int cmd, unsigned long arg)
{
	if (file == NULL)
		return -EINVAL;
	return zvol_compat_ioctl(file->f_dentry->d_inode->i_bdev,
	                         file->f_mode, cmd, arg);
}
# else
# define zvol_compat_ioctl_by_inode   NULL
# endif

static struct block_device_operations zvol_ops = {
	.open            = zvol_open_by_inode,
	.release         = zvol_release_by_inode,
	.ioctl           = zvol_ioctl_by_inode,
	.compat_ioctl    = zvol_compat_ioctl_by_inode,
	.media_changed   = zvol_media_changed,
	.revalidate_disk = zvol_revalidate_disk,
	.getgeo          = zvol_getgeo,
        .owner           = THIS_MODULE,
};
#endif /* HAVE_BDEV_BLOCK_DEVICE_OPERATIONS */

/*
 * Allocate memory for a new zvol_state_t and setup the required
 * request queue and generic disk structures for the block device.
 */
static zvol_state_t *
zvol_alloc(dev_t dev, const char *name)
{
	zvol_state_t *zv;
	int error = 0;

	zv = kmem_zalloc(sizeof (zvol_state_t), KM_SLEEP);

	spin_lock_init(&zv->zv_lock);
	list_link_init(&zv->zv_next);

	zv->zv_queue = blk_init_queue(zvol_request, &zv->zv_lock);
	if (zv->zv_queue == NULL)
		goto out_kmem;

#ifdef HAVE_ELEVATOR_CHANGE
	error = elevator_change(zv->zv_queue, "noop");
#endif /* HAVE_ELEVATOR_CHANGE */
	if (error) {
		printk("ZFS: Unable to set \"%s\" scheduler for zvol %s: %d\n",
		    "noop", name, error);
		goto out_queue;
	}

#ifdef HAVE_BLK_QUEUE_FLUSH
	blk_queue_flush(zv->zv_queue, VDEV_REQ_FLUSH | VDEV_REQ_FUA);
#else
	blk_queue_ordered(zv->zv_queue, QUEUE_ORDERED_DRAIN, NULL);
#endif /* HAVE_BLK_QUEUE_FLUSH */

	zv->zv_disk = alloc_disk(ZVOL_MINORS);
	if (zv->zv_disk == NULL)
		goto out_queue;

	zv->zv_queue->queuedata = zv;
	zv->zv_dev = dev;
	zv->zv_open_count = 0;
	strlcpy(zv->zv_name, name, MAXNAMELEN);

	mutex_init(&zv->zv_znode.z_range_lock, NULL, MUTEX_DEFAULT, NULL);
	avl_create(&zv->zv_znode.z_range_avl, zfs_range_compare,
	    sizeof (rl_t), offsetof(rl_t, r_node));
	zv->zv_znode.z_is_zvol = TRUE;

	zv->zv_disk->major = zvol_major;
	zv->zv_disk->first_minor = (dev & MINORMASK);
	zv->zv_disk->fops = &zvol_ops;
	zv->zv_disk->private_data = zv;
	zv->zv_disk->queue = zv->zv_queue;
	snprintf(zv->zv_disk->disk_name, DISK_NAME_LEN, "%s%d",
	    ZVOL_DEV_NAME, (dev & MINORMASK));

	return zv;

out_queue:
	blk_cleanup_queue(zv->zv_queue);
out_kmem:
	kmem_free(zv, sizeof (zvol_state_t));

	return NULL;
}

/*
 * Cleanup then free a zvol_state_t which was created by zvol_alloc().
 */
static void
zvol_free(zvol_state_t *zv)
{
	avl_destroy(&zv->zv_znode.z_range_avl);
	mutex_destroy(&zv->zv_znode.z_range_lock);

	del_gendisk(zv->zv_disk);
	blk_cleanup_queue(zv->zv_queue);
	put_disk(zv->zv_disk);

	kmem_free(zv, sizeof (zvol_state_t));
}

static int
__zvol_snapdev_hidden(const char *name)
{
        uint64_t snapdev;
        char *parent;
        char *atp;
        int error = 0;

        parent = kmem_alloc(MAXPATHLEN, KM_SLEEP);
        (void) strlcpy(parent, name, MAXPATHLEN);

        if ((atp = strrchr(parent, '@')) != NULL) {
                *atp = '\0';
                error = dsl_prop_get_integer(parent, "snapdev", &snapdev, NULL);
                if ((error == 0) && (snapdev == ZFS_SNAPDEV_HIDDEN))
                        error = SET_ERROR(ENODEV);
        }
        kmem_free(parent, MAXPATHLEN);
        return (error);
}

static int
__zvol_create_minor(const char *name, boolean_t ignore_snapdev)
{
	zvol_state_t *zv;
	objset_t *os;
	dmu_object_info_t *doi;
	uint64_t volsize;
	unsigned minor = 0;
	int error = 0;

	ASSERT(MUTEX_HELD(&zvol_state_lock));

	zv = zvol_find_by_name(name);
	if (zv) {
		error = SET_ERROR(EEXIST);
		goto out;
	}

	if (ignore_snapdev == B_FALSE) {
		error = __zvol_snapdev_hidden(name);
		if (error)
			goto out;
	}

	doi = kmem_alloc(sizeof(dmu_object_info_t), KM_SLEEP);

	error = dmu_objset_own(name, DMU_OST_ZVOL, B_TRUE, zvol_tag, &os);
	if (error)
		goto out_doi;

	error = dmu_object_info(os, ZVOL_OBJ, doi);
	if (error)
		goto out_dmu_objset_disown;

	error = zap_lookup(os, ZVOL_ZAP_OBJ, "size", 8, 1, &volsize);
	if (error)
		goto out_dmu_objset_disown;

	error = zvol_find_minor(&minor);
	if (error)
		goto out_dmu_objset_disown;

	zv = zvol_alloc(MKDEV(zvol_major, minor), name);
	if (zv == NULL) {
		error = SET_ERROR(EAGAIN);
		goto out_dmu_objset_disown;
	}

	if (dmu_objset_is_snapshot(os))
		zv->zv_flags |= ZVOL_RDONLY;

	zv->zv_volblocksize = doi->doi_data_block_size;
	zv->zv_volsize = volsize;
	zv->zv_objset = os;

	set_capacity(zv->zv_disk, zv->zv_volsize >> 9);

	blk_queue_max_hw_sectors(zv->zv_queue, UINT_MAX);
	blk_queue_max_segments(zv->zv_queue, UINT16_MAX);
	blk_queue_max_segment_size(zv->zv_queue, UINT_MAX);
	blk_queue_physical_block_size(zv->zv_queue, zv->zv_volblocksize);
	blk_queue_io_opt(zv->zv_queue, zv->zv_volblocksize);
#ifdef HAVE_BLK_QUEUE_DISCARD
	blk_queue_max_discard_sectors(zv->zv_queue,
	    (zvol_max_discard_blocks * zv->zv_volblocksize) >> 9);
	blk_queue_discard_granularity(zv->zv_queue, zv->zv_volblocksize);
	queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, zv->zv_queue);
#endif
#ifdef HAVE_BLK_QUEUE_NONROT
	queue_flag_set_unlocked(QUEUE_FLAG_NONROT, zv->zv_queue);
#endif

	if (spa_writeable(dmu_objset_spa(os))) {
		if (zil_replay_disable)
			zil_destroy(dmu_objset_zil(os), B_FALSE);
		else
			zil_replay(os, zv, zvol_replay_vector);
	}

	zv->zv_objset = NULL;
out_dmu_objset_disown:
	dmu_objset_disown(os, zvol_tag);
out_doi:
	kmem_free(doi, sizeof(dmu_object_info_t));
out:

	if (error == 0) {
		zvol_insert(zv);
		add_disk(zv->zv_disk);
	}

	return (error);
}

/*
 * Create a block device minor node and setup the linkage between it
 * and the specified volume.  Once this function returns the block
 * device is live and ready for use.
 */
int
zvol_create_minor(const char *name)
{
	int error;

	mutex_enter(&zvol_state_lock);
	error = __zvol_create_minor(name, B_FALSE);
	mutex_exit(&zvol_state_lock);

	return (error);
}

static int
__zvol_remove_minor(const char *name)
{
	zvol_state_t *zv;

	ASSERT(MUTEX_HELD(&zvol_state_lock));

	zv = zvol_find_by_name(name);
	if (zv == NULL)
		return (SET_ERROR(ENXIO));

	if (zv->zv_open_count > 0)
		return (SET_ERROR(EBUSY));

	zvol_remove(zv);
	zvol_free(zv);

	return (0);
}

/*
 * Remove a block device minor node for the specified volume.
 */
int
zvol_remove_minor(const char *name)
{
	int error;

	mutex_enter(&zvol_state_lock);
	error = __zvol_remove_minor(name);
	mutex_exit(&zvol_state_lock);

	return (error);
}

static int
zvol_create_minors_cb(const char *dsname, void *arg)
{
	if (strchr(dsname, '/') == NULL)
		return 0;

	(void) __zvol_create_minor(dsname, B_FALSE);
	return (0);
}

/*
 * Create minors for specified pool, if pool is NULL create minors
 * for all available pools.
 */
int
zvol_create_minors(char *pool)
{
	spa_t *spa = NULL;
	int error = 0;

	if (zvol_inhibit_dev)
		return (0);

	mutex_enter(&zvol_state_lock);
	if (pool) {
		error = dmu_objset_find(pool, zvol_create_minors_cb,
		    NULL, DS_FIND_CHILDREN | DS_FIND_SNAPSHOTS);
	} else {
		mutex_enter(&spa_namespace_lock);
		while ((spa = spa_next(spa)) != NULL) {
			error = dmu_objset_find(spa_name(spa), zvol_create_minors_cb, NULL,
			    DS_FIND_CHILDREN | DS_FIND_SNAPSHOTS);
			if (error)
				break;
		}
		mutex_exit(&spa_namespace_lock);
	}
	mutex_exit(&zvol_state_lock);

	return error;
}

/*
 * Remove minors for specified pool, if pool is NULL remove all minors.
 */
void
zvol_remove_minors(const char *pool)
{
	zvol_state_t *zv, *zv_next;
	char *str;

	if (zvol_inhibit_dev)
		return;

	str = kmem_zalloc(MAXNAMELEN, KM_SLEEP);
	if (pool) {
		(void) strncpy(str, pool, strlen(pool));
		(void) strcat(str, "/");
	}

	mutex_enter(&zvol_state_lock);
	for (zv = list_head(&zvol_state_list); zv != NULL; zv = zv_next) {
		zv_next = list_next(&zvol_state_list, zv);

		if (pool == NULL || !strncmp(str, zv->zv_name, strlen(str))) {
			zvol_remove(zv);
			zvol_free(zv);
		}
	}
	mutex_exit(&zvol_state_lock);
	kmem_free(str, MAXNAMELEN);
}

static int
snapdev_snapshot_changed_cb(const char *dsname, void *arg) {
	uint64_t snapdev = *(uint64_t *) arg;

	if (strchr(dsname, '@') == NULL)
		return 0;

	switch (snapdev) {
		case ZFS_SNAPDEV_VISIBLE:
			mutex_enter(&zvol_state_lock);
			(void) __zvol_create_minor(dsname, B_TRUE);
			mutex_exit(&zvol_state_lock);
			break;
		case ZFS_SNAPDEV_HIDDEN:
			(void) zvol_remove_minor(dsname);
			break;
	}
	return 0;
}

int
zvol_set_snapdev(const char *dsname, uint64_t snapdev) {
	(void) dmu_objset_find((char *) dsname, snapdev_snapshot_changed_cb,
		&snapdev, DS_FIND_SNAPSHOTS | DS_FIND_CHILDREN);
	/* caller should continue to modify snapdev property */
	return (-1);
}


int
zvol_init(void)
{
	int error;

	list_create(&zvol_state_list, sizeof (zvol_state_t),
	            offsetof(zvol_state_t, zv_next));
	mutex_init(&zvol_state_lock, NULL, MUTEX_DEFAULT, NULL);

	zvol_taskq = taskq_create(ZVOL_DRIVER, zvol_threads, maxclsyspri,
		                  zvol_threads, INT_MAX, TASKQ_PREPOPULATE);
	if (zvol_taskq == NULL) {
		printk(KERN_INFO "ZFS: taskq_create() failed\n");
		error = -ENOMEM;
		goto out1;
	}

	error = register_blkdev(zvol_major, ZVOL_DRIVER);
	if (error) {
		printk(KERN_INFO "ZFS: register_blkdev() failed %d\n", error);
		goto out2;
	}

	blk_register_region(MKDEV(zvol_major, 0), 1UL << MINORBITS,
	                    THIS_MODULE, zvol_probe, NULL, NULL);

	return (0);

out2:
	taskq_destroy(zvol_taskq);
out1:
	mutex_destroy(&zvol_state_lock);
	list_destroy(&zvol_state_list);

	return (error);
}

void
zvol_fini(void)
{
	zvol_remove_minors(NULL);
	blk_unregister_region(MKDEV(zvol_major, 0), 1UL << MINORBITS);
	unregister_blkdev(zvol_major, ZVOL_DRIVER);
	taskq_destroy(zvol_taskq);
	mutex_destroy(&zvol_state_lock);
	list_destroy(&zvol_state_list);
}

module_param(zvol_inhibit_dev, uint, 0644);
MODULE_PARM_DESC(zvol_inhibit_dev, "Do not create zvol device nodes");

module_param(zvol_major, uint, 0444);
MODULE_PARM_DESC(zvol_major, "Major number for zvol device");

module_param(zvol_threads, uint, 0444);
MODULE_PARM_DESC(zvol_threads, "Number of threads for zvol device");

module_param(zvol_max_discard_blocks, ulong, 0444);
MODULE_PARM_DESC(zvol_max_discard_blocks, "Max number of blocks to discard at once");