File: zbd.c

package info (click to toggle)
fio 3.12-2
  • links: PTS, VCS
  • area: main
  • in suites: bullseye, buster, sid
  • size: 4,488 kB
  • sloc: ansic: 65,165; sh: 3,284; python: 1,978; makefile: 657; yacc: 204; lex: 184
file content (1339 lines) | stat: -rw-r--r-- 35,984 bytes parent folder | download
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
/*
 * Copyright (C) 2018 Western Digital Corporation or its affiliates.
 *
 * This file is released under the GPL.
 */

#include <errno.h>
#include <string.h>
#include <stdlib.h>
#include <dirent.h>
#include <fcntl.h>
#include <sys/ioctl.h>
#include <sys/stat.h>
#include <unistd.h>
#include <linux/blkzoned.h>
#include "file.h"
#include "fio.h"
#include "lib/pow2.h"
#include "log.h"
#include "smalloc.h"
#include "verify.h"
#include "zbd.h"

/**
 * zbd_zone_idx - convert an offset into a zone number
 * @f: file pointer.
 * @offset: offset in bytes. If this offset is in the first zone_size bytes
 *	    past the disk size then the index of the sentinel is returned.
 */
static uint32_t zbd_zone_idx(const struct fio_file *f, uint64_t offset)
{
	uint32_t zone_idx;

	if (f->zbd_info->zone_size_log2 > 0)
		zone_idx = offset >> f->zbd_info->zone_size_log2;
	else
		zone_idx = offset / f->zbd_info->zone_size;

	return min(zone_idx, f->zbd_info->nr_zones);
}

/**
 * zbd_zone_full - verify whether a minimum number of bytes remain in a zone
 * @f: file pointer.
 * @z: zone info pointer.
 * @required: minimum number of bytes that must remain in a zone.
 *
 * The caller must hold z->mutex.
 */
static bool zbd_zone_full(const struct fio_file *f, struct fio_zone_info *z,
			  uint64_t required)
{
	assert((required & 511) == 0);

	return z->type == BLK_ZONE_TYPE_SEQWRITE_REQ &&
		z->wp + required > z->start + f->zbd_info->zone_size;
}

static bool is_valid_offset(const struct fio_file *f, uint64_t offset)
{
	return (uint64_t)(offset - f->file_offset) < f->io_size;
}

/* Verify whether direct I/O is used for all host-managed zoned drives. */
static bool zbd_using_direct_io(void)
{
	struct thread_data *td;
	struct fio_file *f;
	int i, j;

	for_each_td(td, i) {
		if (td->o.odirect || !(td->o.td_ddir & TD_DDIR_WRITE))
			continue;
		for_each_file(td, f, j) {
			if (f->zbd_info &&
			    f->zbd_info->model == ZBD_DM_HOST_MANAGED)
				return false;
		}
	}

	return true;
}

/* Whether or not the I/O range for f includes one or more sequential zones */
static bool zbd_is_seq_job(struct fio_file *f)
{
	uint32_t zone_idx, zone_idx_b, zone_idx_e;

	assert(f->zbd_info);
	if (f->io_size == 0)
		return false;
	zone_idx_b = zbd_zone_idx(f, f->file_offset);
	zone_idx_e = zbd_zone_idx(f, f->file_offset + f->io_size - 1);
	for (zone_idx = zone_idx_b; zone_idx <= zone_idx_e; zone_idx++)
		if (f->zbd_info->zone_info[zone_idx].type ==
		    BLK_ZONE_TYPE_SEQWRITE_REQ)
			return true;

	return false;
}

/*
 * Verify whether offset and size parameters are aligned with zone boundaries.
 */
static bool zbd_verify_sizes(void)
{
	const struct fio_zone_info *z;
	struct thread_data *td;
	struct fio_file *f;
	uint64_t new_offset, new_end;
	uint32_t zone_idx;
	int i, j;

	for_each_td(td, i) {
		for_each_file(td, f, j) {
			if (!f->zbd_info)
				continue;
			if (f->file_offset >= f->real_file_size)
				continue;
			if (!zbd_is_seq_job(f))
				continue;
			zone_idx = zbd_zone_idx(f, f->file_offset);
			z = &f->zbd_info->zone_info[zone_idx];
			if (f->file_offset != z->start) {
				new_offset = (z+1)->start;
				if (new_offset >= f->file_offset + f->io_size) {
					log_info("%s: io_size must be at least one zone\n",
						 f->file_name);
					return false;
				}
				log_info("%s: rounded up offset from %llu to %llu\n",
					 f->file_name, (unsigned long long) f->file_offset,
					 (unsigned long long) new_offset);
				f->io_size -= (new_offset - f->file_offset);
				f->file_offset = new_offset;
			}
			zone_idx = zbd_zone_idx(f, f->file_offset + f->io_size);
			z = &f->zbd_info->zone_info[zone_idx];
			new_end = z->start;
			if (f->file_offset + f->io_size != new_end) {
				if (new_end <= f->file_offset) {
					log_info("%s: io_size must be at least one zone\n",
						 f->file_name);
					return false;
				}
				log_info("%s: rounded down io_size from %llu to %llu\n",
					 f->file_name, (unsigned long long) f->io_size,
					 (unsigned long long) new_end - f->file_offset);
				f->io_size = new_end - f->file_offset;
			}
		}
	}

	return true;
}

static bool zbd_verify_bs(void)
{
	struct thread_data *td;
	struct fio_file *f;
	uint32_t zone_size;
	int i, j, k;

	for_each_td(td, i) {
		for_each_file(td, f, j) {
			if (!f->zbd_info)
				continue;
			zone_size = f->zbd_info->zone_size;
			for (k = 0; k < ARRAY_SIZE(td->o.bs); k++) {
				if (td->o.verify != VERIFY_NONE &&
				    zone_size % td->o.bs[k] != 0) {
					log_info("%s: block size %llu is not a divisor of the zone size %d\n",
						 f->file_name, td->o.bs[k],
						 zone_size);
					return false;
				}
			}
		}
	}
	return true;
}

/*
 * Read zone information into @buf starting from sector @start_sector.
 * @fd is a file descriptor that refers to a block device and @bufsz is the
 * size of @buf.
 *
 * Returns 0 upon success and a negative error code upon failure.
 */
static int read_zone_info(int fd, uint64_t start_sector,
			  void *buf, unsigned int bufsz)
{
	struct blk_zone_report *hdr = buf;

	if (bufsz < sizeof(*hdr))
		return -EINVAL;

	memset(hdr, 0, sizeof(*hdr));

	hdr->nr_zones = (bufsz - sizeof(*hdr)) / sizeof(struct blk_zone);
	hdr->sector = start_sector;
	return ioctl(fd, BLKREPORTZONE, hdr) >= 0 ? 0 : -errno;
}

/*
 * Read up to 255 characters from the first line of a file. Strip the trailing
 * newline.
 */
static char *read_file(const char *path)
{
	char line[256], *p = line;
	FILE *f;

	f = fopen(path, "rb");
	if (!f)
		return NULL;
	if (!fgets(line, sizeof(line), f))
		line[0] = '\0';
	strsep(&p, "\n");
	fclose(f);

	return strdup(line);
}

static enum blk_zoned_model get_zbd_model(const char *file_name)
{
	enum blk_zoned_model model = ZBD_DM_NONE;
	char *zoned_attr_path = NULL;
	char *model_str = NULL;
	struct stat statbuf;

	if (stat(file_name, &statbuf) < 0)
		goto out;
	if (asprintf(&zoned_attr_path, "/sys/dev/block/%d:%d/queue/zoned",
		     major(statbuf.st_rdev), minor(statbuf.st_rdev)) < 0)
		goto out;
	model_str = read_file(zoned_attr_path);
	if (!model_str)
		goto out;
	dprint(FD_ZBD, "%s: zbd model string: %s\n", file_name, model_str);
	if (strcmp(model_str, "host-aware") == 0)
		model = ZBD_DM_HOST_AWARE;
	else if (strcmp(model_str, "host-managed") == 0)
		model = ZBD_DM_HOST_MANAGED;

out:
	free(model_str);
	free(zoned_attr_path);
	return model;
}

static int ilog2(uint64_t i)
{
	int log = -1;

	while (i) {
		i >>= 1;
		log++;
	}
	return log;
}

/*
 * Initialize f->zbd_info for devices that are not zoned block devices. This
 * allows to execute a ZBD workload against a non-ZBD device.
 */
static int init_zone_info(struct thread_data *td, struct fio_file *f)
{
	uint32_t nr_zones;
	struct fio_zone_info *p;
	uint64_t zone_size;
	struct zoned_block_device_info *zbd_info = NULL;
	pthread_mutexattr_t attr;
	int i;

	zone_size = td->o.zone_size;
	assert(zone_size);
	nr_zones = (f->real_file_size + zone_size - 1) / zone_size;
	zbd_info = scalloc(1, sizeof(*zbd_info) +
			   (nr_zones + 1) * sizeof(zbd_info->zone_info[0]));
	if (!zbd_info)
		return -ENOMEM;

	pthread_mutexattr_init(&attr);
	pthread_mutexattr_settype(&attr, PTHREAD_MUTEX_RECURSIVE);
	pthread_mutexattr_setpshared(&attr, true);
	pthread_mutex_init(&zbd_info->mutex, &attr);
	zbd_info->refcount = 1;
	p = &zbd_info->zone_info[0];
	for (i = 0; i < nr_zones; i++, p++) {
		pthread_mutex_init(&p->mutex, &attr);
		p->start = i * zone_size;
		p->wp = p->start + zone_size;
		p->type = BLK_ZONE_TYPE_SEQWRITE_REQ;
		p->cond = BLK_ZONE_COND_EMPTY;
	}
	/* a sentinel */
	p->start = nr_zones * zone_size;

	f->zbd_info = zbd_info;
	f->zbd_info->zone_size = zone_size;
	f->zbd_info->zone_size_log2 = is_power_of_2(zone_size) ?
		ilog2(zone_size) : -1;
	f->zbd_info->nr_zones = nr_zones;
	pthread_mutexattr_destroy(&attr);
	return 0;
}

/*
 * Parse the BLKREPORTZONE output and store it in f->zbd_info. Must be called
 * only for devices that support this ioctl, namely zoned block devices.
 */
static int parse_zone_info(struct thread_data *td, struct fio_file *f)
{
	const unsigned int bufsz = sizeof(struct blk_zone_report) +
		4096 * sizeof(struct blk_zone);
	uint32_t nr_zones;
	struct blk_zone_report *hdr;
	const struct blk_zone *z;
	struct fio_zone_info *p;
	uint64_t zone_size, start_sector;
	struct zoned_block_device_info *zbd_info = NULL;
	pthread_mutexattr_t attr;
	void *buf;
	int fd, i, j, ret = 0;

	pthread_mutexattr_init(&attr);
	pthread_mutexattr_settype(&attr, PTHREAD_MUTEX_RECURSIVE);
	pthread_mutexattr_setpshared(&attr, true);

	buf = malloc(bufsz);
	if (!buf)
		goto out;

	fd = open(f->file_name, O_RDONLY | O_LARGEFILE);
	if (fd < 0) {
		ret = -errno;
		goto free;
	}

	ret = read_zone_info(fd, 0, buf, bufsz);
	if (ret < 0) {
		log_info("fio: BLKREPORTZONE(%lu) failed for %s (%d).\n",
			 0UL, f->file_name, -ret);
		goto close;
	}
	hdr = buf;
	if (hdr->nr_zones < 1) {
		log_info("fio: %s has invalid zone information.\n",
			 f->file_name);
		goto close;
	}
	z = (void *)(hdr + 1);
	zone_size = z->len << 9;
	nr_zones = (f->real_file_size + zone_size - 1) / zone_size;

	if (td->o.zone_size == 0) {
		td->o.zone_size = zone_size;
	} else if (td->o.zone_size != zone_size) {
		log_info("fio: %s job parameter zonesize %llu does not match disk zone size %llu.\n",
			 f->file_name, (unsigned long long) td->o.zone_size,
			(unsigned long long) zone_size);
		ret = -EINVAL;
		goto close;
	}

	dprint(FD_ZBD, "Device %s has %d zones of size %llu KB\n", f->file_name,
	       nr_zones, (unsigned long long) zone_size / 1024);

	zbd_info = scalloc(1, sizeof(*zbd_info) +
			   (nr_zones + 1) * sizeof(zbd_info->zone_info[0]));
	ret = -ENOMEM;
	if (!zbd_info)
		goto close;
	pthread_mutex_init(&zbd_info->mutex, &attr);
	zbd_info->refcount = 1;
	p = &zbd_info->zone_info[0];
	for (start_sector = 0, j = 0; j < nr_zones;) {
		z = (void *)(hdr + 1);
		for (i = 0; i < hdr->nr_zones; i++, j++, z++, p++) {
			pthread_mutex_init(&p->mutex, &attr);
			p->start = z->start << 9;
			switch (z->cond) {
			case BLK_ZONE_COND_NOT_WP:
				p->wp = p->start;
				break;
			case BLK_ZONE_COND_FULL:
				p->wp = p->start + zone_size;
				break;
			default:
				assert(z->start <= z->wp);
				assert(z->wp <= z->start + (zone_size >> 9));
				p->wp = z->wp << 9;
				break;
			}
			p->type = z->type;
			p->cond = z->cond;
			if (j > 0 && p->start != p[-1].start + zone_size) {
				log_info("%s: invalid zone data\n",
					 f->file_name);
				ret = -EINVAL;
				goto close;
			}
		}
		z--;
		start_sector = z->start + z->len;
		if (j >= nr_zones)
			break;
		ret = read_zone_info(fd, start_sector, buf, bufsz);
		if (ret < 0) {
			log_info("fio: BLKREPORTZONE(%llu) failed for %s (%d).\n",
				 (unsigned long long) start_sector, f->file_name, -ret);
			goto close;
		}
	}
	/* a sentinel */
	zbd_info->zone_info[nr_zones].start = start_sector << 9;

	f->zbd_info = zbd_info;
	f->zbd_info->zone_size = zone_size;
	f->zbd_info->zone_size_log2 = is_power_of_2(zone_size) ?
		ilog2(zone_size) : -1;
	f->zbd_info->nr_zones = nr_zones;
	zbd_info = NULL;
	ret = 0;

close:
	sfree(zbd_info);
	close(fd);
free:
	free(buf);
out:
	pthread_mutexattr_destroy(&attr);
	return ret;
}

/*
 * Allocate zone information and store it into f->zbd_info if zonemode=zbd.
 *
 * Returns 0 upon success and a negative error code upon failure.
 */
int zbd_create_zone_info(struct thread_data *td, struct fio_file *f)
{
	enum blk_zoned_model zbd_model;
	int ret = 0;

	assert(td->o.zone_mode == ZONE_MODE_ZBD);

	zbd_model = get_zbd_model(f->file_name);
	switch (zbd_model) {
	case ZBD_DM_HOST_AWARE:
	case ZBD_DM_HOST_MANAGED:
		ret = parse_zone_info(td, f);
		break;
	case ZBD_DM_NONE:
		ret = init_zone_info(td, f);
		break;
	}
	if (ret == 0)
		f->zbd_info->model = zbd_model;
	return ret;
}

void zbd_free_zone_info(struct fio_file *f)
{
	uint32_t refcount;

	if (!f->zbd_info)
		return;

	pthread_mutex_lock(&f->zbd_info->mutex);
	refcount = --f->zbd_info->refcount;
	pthread_mutex_unlock(&f->zbd_info->mutex);

	assert((int32_t)refcount >= 0);
	if (refcount == 0)
		sfree(f->zbd_info);
	f->zbd_info = NULL;
}

/*
 * Initialize f->zbd_info.
 *
 * Returns 0 upon success and a negative error code upon failure.
 *
 * Note: this function can only work correctly if it is called before the first
 * fio fork() call.
 */
static int zbd_init_zone_info(struct thread_data *td, struct fio_file *file)
{
	struct thread_data *td2;
	struct fio_file *f2;
	int i, j, ret;

	for_each_td(td2, i) {
		for_each_file(td2, f2, j) {
			if (td2 == td && f2 == file)
				continue;
			if (!f2->zbd_info ||
			    strcmp(f2->file_name, file->file_name) != 0)
				continue;
			file->zbd_info = f2->zbd_info;
			file->zbd_info->refcount++;
			return 0;
		}
	}

	ret = zbd_create_zone_info(td, file);
	if (ret < 0)
		td_verror(td, -ret, "BLKREPORTZONE failed");
	return ret;
}

int zbd_init(struct thread_data *td)
{
	struct fio_file *f;
	int i;

	for_each_file(td, f, i) {
		if (f->filetype != FIO_TYPE_BLOCK)
			continue;
		if (td->o.zone_size && td->o.zone_size < 512) {
			log_err("%s: zone size must be at least 512 bytes for --zonemode=zbd\n\n",
				f->file_name);
			return 1;
		}
		if (td->o.zone_size == 0 &&
		    get_zbd_model(f->file_name) == ZBD_DM_NONE) {
			log_err("%s: Specifying the zone size is mandatory for regular block devices with --zonemode=zbd\n\n",
				f->file_name);
			return 1;
		}
		zbd_init_zone_info(td, f);
	}

	if (!zbd_using_direct_io()) {
		log_err("Using direct I/O is mandatory for writing to ZBD drives\n\n");
		return 1;
	}

	if (!zbd_verify_sizes())
		return 1;

	if (!zbd_verify_bs())
		return 1;

	return 0;
}

/**
 * zbd_reset_range - reset zones for a range of sectors
 * @td: FIO thread data.
 * @f: Fio file for which to reset zones
 * @sector: Starting sector in units of 512 bytes
 * @nr_sectors: Number of sectors in units of 512 bytes
 *
 * Returns 0 upon success and a negative error code upon failure.
 */
static int zbd_reset_range(struct thread_data *td, const struct fio_file *f,
			   uint64_t offset, uint64_t length)
{
	struct blk_zone_range zr = {
		.sector         = offset >> 9,
		.nr_sectors     = length >> 9,
	};
	uint32_t zone_idx_b, zone_idx_e;
	struct fio_zone_info *zb, *ze, *z;
	int ret = 0;

	assert(f->fd != -1);
	assert(is_valid_offset(f, offset + length - 1));
	switch (f->zbd_info->model) {
	case ZBD_DM_HOST_AWARE:
	case ZBD_DM_HOST_MANAGED:
		ret = ioctl(f->fd, BLKRESETZONE, &zr);
		if (ret < 0) {
			td_verror(td, errno, "resetting wp failed");
			log_err("%s: resetting wp for %llu sectors at sector %llu failed (%d).\n",
				f->file_name, zr.nr_sectors, zr.sector, errno);
			return ret;
		}
		break;
	case ZBD_DM_NONE:
		break;
	}

	zone_idx_b = zbd_zone_idx(f, offset);
	zb = &f->zbd_info->zone_info[zone_idx_b];
	zone_idx_e = zbd_zone_idx(f, offset + length);
	ze = &f->zbd_info->zone_info[zone_idx_e];
	for (z = zb; z < ze; z++) {
		pthread_mutex_lock(&z->mutex);
		pthread_mutex_lock(&f->zbd_info->mutex);
		f->zbd_info->sectors_with_data -= z->wp - z->start;
		pthread_mutex_unlock(&f->zbd_info->mutex);
		z->wp = z->start;
		z->verify_block = 0;
		pthread_mutex_unlock(&z->mutex);
	}

	td->ts.nr_zone_resets += ze - zb;

	return ret;
}

static unsigned int zbd_zone_nr(struct zoned_block_device_info *zbd_info,
				struct fio_zone_info *zone)
{
	return zone - zbd_info->zone_info;
}

/**
 * zbd_reset_zone - reset the write pointer of a single zone
 * @td: FIO thread data.
 * @f: FIO file associated with the disk for which to reset a write pointer.
 * @z: Zone to reset.
 *
 * Returns 0 upon success and a negative error code upon failure.
 */
static int zbd_reset_zone(struct thread_data *td, const struct fio_file *f,
			  struct fio_zone_info *z)
{
	dprint(FD_ZBD, "%s: resetting wp of zone %u.\n", f->file_name,
		zbd_zone_nr(f->zbd_info, z));

	return zbd_reset_range(td, f, z->start, (z+1)->start - z->start);
}

/*
 * Reset a range of zones. Returns 0 upon success and 1 upon failure.
 * @td: fio thread data.
 * @f: fio file for which to reset zones
 * @zb: first zone to reset.
 * @ze: first zone not to reset.
 * @all_zones: whether to reset all zones or only those zones for which the
 *	write pointer is not a multiple of td->o.min_bs[DDIR_WRITE].
 */
static int zbd_reset_zones(struct thread_data *td, struct fio_file *f,
			   struct fio_zone_info *const zb,
			   struct fio_zone_info *const ze, bool all_zones)
{
	struct fio_zone_info *z, *start_z = ze;
	const uint32_t min_bs = td->o.min_bs[DDIR_WRITE];
	bool reset_wp;
	int res = 0;

	dprint(FD_ZBD, "%s: examining zones %u .. %u\n", f->file_name,
		zbd_zone_nr(f->zbd_info, zb), zbd_zone_nr(f->zbd_info, ze));
	assert(f->fd != -1);
	for (z = zb; z < ze; z++) {
		pthread_mutex_lock(&z->mutex);
		switch (z->type) {
		case BLK_ZONE_TYPE_SEQWRITE_REQ:
			reset_wp = all_zones ? z->wp != z->start :
					(td->o.td_ddir & TD_DDIR_WRITE) &&
					z->wp % min_bs != 0;
			if (start_z == ze && reset_wp) {
				start_z = z;
			} else if (start_z < ze && !reset_wp) {
				dprint(FD_ZBD,
				       "%s: resetting zones %u .. %u\n",
				       f->file_name,
					zbd_zone_nr(f->zbd_info, start_z),
					zbd_zone_nr(f->zbd_info, z));
				if (zbd_reset_range(td, f, start_z->start,
						z->start - start_z->start) < 0)
					res = 1;
				start_z = ze;
			}
			break;
		default:
			if (start_z == ze)
				break;
			dprint(FD_ZBD, "%s: resetting zones %u .. %u\n",
			       f->file_name, zbd_zone_nr(f->zbd_info, start_z),
			       zbd_zone_nr(f->zbd_info, z));
			if (zbd_reset_range(td, f, start_z->start,
					    z->start - start_z->start) < 0)
				res = 1;
			start_z = ze;
			break;
		}
	}
	if (start_z < ze) {
		dprint(FD_ZBD, "%s: resetting zones %u .. %u\n", f->file_name,
			zbd_zone_nr(f->zbd_info, start_z),
			zbd_zone_nr(f->zbd_info, z));
		if (zbd_reset_range(td, f, start_z->start,
				    z->start - start_z->start) < 0)
			res = 1;
	}
	for (z = zb; z < ze; z++)
		pthread_mutex_unlock(&z->mutex);

	return res;
}

/*
 * Reset zbd_info.write_cnt, the counter that counts down towards the next
 * zone reset.
 */
static void zbd_reset_write_cnt(const struct thread_data *td,
				const struct fio_file *f)
{
	assert(0 <= td->o.zrf.u.f && td->o.zrf.u.f <= 1);

	pthread_mutex_lock(&f->zbd_info->mutex);
	f->zbd_info->write_cnt = td->o.zrf.u.f ?
		min(1.0 / td->o.zrf.u.f, 0.0 + UINT_MAX) : UINT_MAX;
	pthread_mutex_unlock(&f->zbd_info->mutex);
}

static bool zbd_dec_and_reset_write_cnt(const struct thread_data *td,
					const struct fio_file *f)
{
	uint32_t write_cnt = 0;

	pthread_mutex_lock(&f->zbd_info->mutex);
	assert(f->zbd_info->write_cnt);
	if (f->zbd_info->write_cnt)
		write_cnt = --f->zbd_info->write_cnt;
	if (write_cnt == 0)
		zbd_reset_write_cnt(td, f);
	pthread_mutex_unlock(&f->zbd_info->mutex);

	return write_cnt == 0;
}

enum swd_action {
	CHECK_SWD,
	SET_SWD,
};

/* Calculate the number of sectors with data (swd) and perform action 'a' */
static uint64_t zbd_process_swd(const struct fio_file *f, enum swd_action a)
{
	struct fio_zone_info *zb, *ze, *z;
	uint64_t swd = 0;

	zb = &f->zbd_info->zone_info[zbd_zone_idx(f, f->file_offset)];
	ze = &f->zbd_info->zone_info[zbd_zone_idx(f, f->file_offset +
						  f->io_size)];
	for (z = zb; z < ze; z++) {
		pthread_mutex_lock(&z->mutex);
		swd += z->wp - z->start;
	}
	pthread_mutex_lock(&f->zbd_info->mutex);
	switch (a) {
	case CHECK_SWD:
		assert(f->zbd_info->sectors_with_data == swd);
		break;
	case SET_SWD:
		f->zbd_info->sectors_with_data = swd;
		break;
	}
	pthread_mutex_unlock(&f->zbd_info->mutex);
	for (z = zb; z < ze; z++)
		pthread_mutex_unlock(&z->mutex);

	return swd;
}

/*
 * The swd check is useful for debugging but takes too much time to leave
 * it enabled all the time. Hence it is disabled by default.
 */
static const bool enable_check_swd = false;

/* Check whether the value of zbd_info.sectors_with_data is correct. */
static void zbd_check_swd(const struct fio_file *f)
{
	if (!enable_check_swd)
		return;

	zbd_process_swd(f, CHECK_SWD);
}

static void zbd_init_swd(struct fio_file *f)
{
	uint64_t swd;

	swd = zbd_process_swd(f, SET_SWD);
	dprint(FD_ZBD, "%s(%s): swd = %" PRIu64 "\n", __func__, f->file_name,
	       swd);
}

void zbd_file_reset(struct thread_data *td, struct fio_file *f)
{
	struct fio_zone_info *zb, *ze;
	uint32_t zone_idx_e;

	if (!f->zbd_info)
		return;

	zb = &f->zbd_info->zone_info[zbd_zone_idx(f, f->file_offset)];
	zone_idx_e = zbd_zone_idx(f, f->file_offset + f->io_size);
	ze = &f->zbd_info->zone_info[zone_idx_e];
	zbd_init_swd(f);
	/*
	 * If data verification is enabled reset the affected zones before
	 * writing any data to avoid that a zone reset has to be issued while
	 * writing data, which causes data loss.
	 */
	zbd_reset_zones(td, f, zb, ze, td->o.verify != VERIFY_NONE &&
			(td->o.td_ddir & TD_DDIR_WRITE) &&
			td->runstate != TD_VERIFYING);
	zbd_reset_write_cnt(td, f);
}

/* The caller must hold f->zbd_info->mutex. */
static bool is_zone_open(const struct thread_data *td, const struct fio_file *f,
			 unsigned int zone_idx)
{
	struct zoned_block_device_info *zbdi = f->zbd_info;
	int i;

	assert(td->o.max_open_zones <= ARRAY_SIZE(zbdi->open_zones));
	assert(zbdi->num_open_zones <= td->o.max_open_zones);

	for (i = 0; i < zbdi->num_open_zones; i++)
		if (zbdi->open_zones[i] == zone_idx)
			return true;

	return false;
}

/*
 * Open a ZBD zone if it was not yet open. Returns true if either the zone was
 * already open or if opening a new zone is allowed. Returns false if the zone
 * was not yet open and opening a new zone would cause the zone limit to be
 * exceeded.
 */
static bool zbd_open_zone(struct thread_data *td, const struct io_u *io_u,
			  uint32_t zone_idx)
{
	const uint32_t min_bs = td->o.min_bs[DDIR_WRITE];
	const struct fio_file *f = io_u->file;
	struct fio_zone_info *z = &f->zbd_info->zone_info[zone_idx];
	bool res = true;

	if (z->cond == BLK_ZONE_COND_OFFLINE)
		return false;

	/*
	 * Skip full zones with data verification enabled because resetting a
	 * zone causes data loss and hence causes verification to fail.
	 */
	if (td->o.verify != VERIFY_NONE && zbd_zone_full(f, z, min_bs))
		return false;

	/* Zero means no limit */
	if (!td->o.max_open_zones)
		return true;

	pthread_mutex_lock(&f->zbd_info->mutex);
	if (is_zone_open(td, f, zone_idx))
		goto out;
	res = false;
	if (f->zbd_info->num_open_zones >= td->o.max_open_zones)
		goto out;
	dprint(FD_ZBD, "%s: opening zone %d\n", f->file_name, zone_idx);
	f->zbd_info->open_zones[f->zbd_info->num_open_zones++] = zone_idx;
	z->open = 1;
	res = true;

out:
	pthread_mutex_unlock(&f->zbd_info->mutex);
	return res;
}

/* The caller must hold f->zbd_info->mutex */
static void zbd_close_zone(struct thread_data *td, const struct fio_file *f,
			   unsigned int open_zone_idx)
{
	uint32_t zone_idx;

	assert(open_zone_idx < f->zbd_info->num_open_zones);
	zone_idx = f->zbd_info->open_zones[open_zone_idx];
	memmove(f->zbd_info->open_zones + open_zone_idx,
		f->zbd_info->open_zones + open_zone_idx + 1,
		(FIO_MAX_OPEN_ZBD_ZONES - (open_zone_idx + 1)) *
		sizeof(f->zbd_info->open_zones[0]));
	f->zbd_info->num_open_zones--;
	f->zbd_info->zone_info[zone_idx].open = 0;
}

/*
 * Modify the offset of an I/O unit that does not refer to an open zone such
 * that it refers to an open zone. Close an open zone and open a new zone if
 * necessary. This algorithm can only work correctly if all write pointers are
 * a multiple of the fio block size. The caller must neither hold z->mutex
 * nor f->zbd_info->mutex. Returns with z->mutex held upon success.
 */
struct fio_zone_info *zbd_convert_to_open_zone(struct thread_data *td,
					       struct io_u *io_u)
{
	const uint32_t min_bs = td->o.min_bs[io_u->ddir];
	const struct fio_file *f = io_u->file;
	struct fio_zone_info *z;
	unsigned int open_zone_idx = -1;
	uint32_t zone_idx, new_zone_idx;
	int i;

	assert(is_valid_offset(f, io_u->offset));

	if (td->o.max_open_zones) {
		/*
		 * This statement accesses f->zbd_info->open_zones[] on purpose
		 * without locking.
		 */
		zone_idx = f->zbd_info->open_zones[(io_u->offset -
						    f->file_offset) *
				f->zbd_info->num_open_zones / f->io_size];
	} else {
		zone_idx = zbd_zone_idx(f, io_u->offset);
	}
	dprint(FD_ZBD, "%s(%s): starting from zone %d (offset %lld, buflen %lld)\n",
	       __func__, f->file_name, zone_idx, io_u->offset, io_u->buflen);

	/*
	 * Since z->mutex is the outer lock and f->zbd_info->mutex the inner
	 * lock it can happen that the state of the zone with index zone_idx
	 * has changed after 'z' has been assigned and before f->zbd_info->mutex
	 * has been obtained. Hence the loop.
	 */
	for (;;) {
		z = &f->zbd_info->zone_info[zone_idx];

		pthread_mutex_lock(&z->mutex);
		pthread_mutex_lock(&f->zbd_info->mutex);
		if (td->o.max_open_zones == 0)
			goto examine_zone;
		if (f->zbd_info->num_open_zones == 0) {
			pthread_mutex_unlock(&f->zbd_info->mutex);
			pthread_mutex_unlock(&z->mutex);
			dprint(FD_ZBD, "%s(%s): no zones are open\n",
			       __func__, f->file_name);
			return NULL;
		}
		open_zone_idx = (io_u->offset - f->file_offset) *
			f->zbd_info->num_open_zones / f->io_size;
		assert(open_zone_idx < f->zbd_info->num_open_zones);
		new_zone_idx = f->zbd_info->open_zones[open_zone_idx];
		if (new_zone_idx == zone_idx)
			break;
		zone_idx = new_zone_idx;
		pthread_mutex_unlock(&f->zbd_info->mutex);
		pthread_mutex_unlock(&z->mutex);
	}

	/* Both z->mutex and f->zbd_info->mutex are held. */

examine_zone:
	if (z->wp + min_bs <= (z+1)->start) {
		pthread_mutex_unlock(&f->zbd_info->mutex);
		goto out;
	}
	dprint(FD_ZBD, "%s(%s): closing zone %d\n", __func__, f->file_name,
	       zone_idx);
	if (td->o.max_open_zones)
		zbd_close_zone(td, f, open_zone_idx);
	pthread_mutex_unlock(&f->zbd_info->mutex);

	/* Only z->mutex is held. */

	/* Zone 'z' is full, so try to open a new zone. */
	for (i = f->io_size / f->zbd_info->zone_size; i > 0; i--) {
		zone_idx++;
		pthread_mutex_unlock(&z->mutex);
		z++;
		if (!is_valid_offset(f, z->start)) {
			/* Wrap-around. */
			zone_idx = zbd_zone_idx(f, f->file_offset);
			z = &f->zbd_info->zone_info[zone_idx];
		}
		assert(is_valid_offset(f, z->start));
		pthread_mutex_lock(&z->mutex);
		if (z->open)
			continue;
		if (zbd_open_zone(td, io_u, zone_idx))
			goto out;
	}

	/* Only z->mutex is held. */

	/* Check whether the write fits in any of the already opened zones. */
	pthread_mutex_lock(&f->zbd_info->mutex);
	for (i = 0; i < f->zbd_info->num_open_zones; i++) {
		zone_idx = f->zbd_info->open_zones[i];
		pthread_mutex_unlock(&f->zbd_info->mutex);
		pthread_mutex_unlock(&z->mutex);

		z = &f->zbd_info->zone_info[zone_idx];

		pthread_mutex_lock(&z->mutex);
		if (z->wp + min_bs <= (z+1)->start)
			goto out;
		pthread_mutex_lock(&f->zbd_info->mutex);
	}
	pthread_mutex_unlock(&f->zbd_info->mutex);
	pthread_mutex_unlock(&z->mutex);
	dprint(FD_ZBD, "%s(%s): did not open another zone\n", __func__,
	       f->file_name);
	return NULL;

out:
	dprint(FD_ZBD, "%s(%s): returning zone %d\n", __func__, f->file_name,
	       zone_idx);
	io_u->offset = z->start;
	return z;
}

/* The caller must hold z->mutex. */
static struct fio_zone_info *zbd_replay_write_order(struct thread_data *td,
						    struct io_u *io_u,
						    struct fio_zone_info *z)
{
	const struct fio_file *f = io_u->file;
	const uint32_t min_bs = td->o.min_bs[DDIR_WRITE];

	if (!zbd_open_zone(td, io_u, z - f->zbd_info->zone_info)) {
		pthread_mutex_unlock(&z->mutex);
		z = zbd_convert_to_open_zone(td, io_u);
		assert(z);
	}

	if (z->verify_block * min_bs >= f->zbd_info->zone_size)
		log_err("%s: %d * %d >= %llu\n", f->file_name, z->verify_block,
			min_bs, (unsigned long long) f->zbd_info->zone_size);
	io_u->offset = z->start + z->verify_block++ * min_bs;
	return z;
}

/*
 * Find another zone for which @io_u fits below the write pointer. Start
 * searching in zones @zb + 1 .. @zl and continue searching in zones
 * @zf .. @zb - 1.
 *
 * Either returns NULL or returns a zone pointer and holds the mutex for that
 * zone.
 */
static struct fio_zone_info *
zbd_find_zone(struct thread_data *td, struct io_u *io_u,
	      struct fio_zone_info *zb, struct fio_zone_info *zl)
{
	const uint32_t min_bs = td->o.min_bs[io_u->ddir];
	const struct fio_file *f = io_u->file;
	struct fio_zone_info *z1, *z2;
	const struct fio_zone_info *const zf =
		&f->zbd_info->zone_info[zbd_zone_idx(f, f->file_offset)];

	/*
	 * Skip to the next non-empty zone in case of sequential I/O and to
	 * the nearest non-empty zone in case of random I/O.
	 */
	for (z1 = zb + 1, z2 = zb - 1; z1 < zl || z2 >= zf; z1++, z2--) {
		if (z1 < zl && z1->cond != BLK_ZONE_COND_OFFLINE) {
			pthread_mutex_lock(&z1->mutex);
			if (z1->start + min_bs <= z1->wp)
				return z1;
			pthread_mutex_unlock(&z1->mutex);
		} else if (!td_random(td)) {
			break;
		}
		if (td_random(td) && z2 >= zf &&
		    z2->cond != BLK_ZONE_COND_OFFLINE) {
			pthread_mutex_lock(&z2->mutex);
			if (z2->start + min_bs <= z2->wp)
				return z2;
			pthread_mutex_unlock(&z2->mutex);
		}
	}
	dprint(FD_ZBD, "%s: adjusting random read offset failed\n",
	       f->file_name);
	return NULL;
}


/**
 * zbd_post_submit - update the write pointer and unlock the zone lock
 * @io_u: I/O unit
 * @success: Whether or not the I/O unit has been executed successfully
 *
 * For write and trim operations, update the write pointer of all affected
 * zones.
 */
static void zbd_post_submit(const struct io_u *io_u, bool success)
{
	struct zoned_block_device_info *zbd_info;
	struct fio_zone_info *z;
	uint32_t zone_idx;
	uint64_t end, zone_end;

	zbd_info = io_u->file->zbd_info;
	if (!zbd_info)
		return;

	zone_idx = zbd_zone_idx(io_u->file, io_u->offset);
	end = io_u->offset + io_u->buflen;
	z = &zbd_info->zone_info[zone_idx];
	assert(zone_idx < zbd_info->nr_zones);
	if (z->type != BLK_ZONE_TYPE_SEQWRITE_REQ)
		return;
	if (!success)
		goto unlock;
	switch (io_u->ddir) {
	case DDIR_WRITE:
		zone_end = min(end, (z + 1)->start);
		pthread_mutex_lock(&zbd_info->mutex);
		/*
		 * z->wp > zone_end means that one or more I/O errors
		 * have occurred.
		 */
		if (z->wp <= zone_end)
			zbd_info->sectors_with_data += zone_end - z->wp;
		pthread_mutex_unlock(&zbd_info->mutex);
		z->wp = zone_end;
		break;
	case DDIR_TRIM:
		assert(z->wp == z->start);
		break;
	default:
		break;
	}
unlock:
	pthread_mutex_unlock(&z->mutex);

	zbd_check_swd(io_u->file);
}

bool zbd_unaligned_write(int error_code)
{
	switch (error_code) {
	case EIO:
	case EREMOTEIO:
		return true;
	}
	return false;
}

/**
 * zbd_adjust_block - adjust the offset and length as necessary for ZBD drives
 * @td: FIO thread data.
 * @io_u: FIO I/O unit.
 *
 * Locking strategy: returns with z->mutex locked if and only if z refers
 * to a sequential zone and if io_u_accept is returned. z is the zone that
 * corresponds to io_u->offset at the end of this function.
 */
enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
{
	const struct fio_file *f = io_u->file;
	uint32_t zone_idx_b;
	struct fio_zone_info *zb, *zl, *orig_zb;
	uint32_t orig_len = io_u->buflen;
	uint32_t min_bs = td->o.min_bs[io_u->ddir];
	uint64_t new_len;
	int64_t range;

	if (!f->zbd_info)
		return io_u_accept;

	assert(is_valid_offset(f, io_u->offset));
	assert(io_u->buflen);
	zone_idx_b = zbd_zone_idx(f, io_u->offset);
	zb = &f->zbd_info->zone_info[zone_idx_b];
	orig_zb = zb;

	/* Accept the I/O offset for conventional zones. */
	if (zb->type == BLK_ZONE_TYPE_CONVENTIONAL)
		return io_u_accept;

	/*
	 * Accept the I/O offset for reads if reading beyond the write pointer
	 * is enabled.
	 */
	if (zb->cond != BLK_ZONE_COND_OFFLINE &&
	    io_u->ddir == DDIR_READ && td->o.read_beyond_wp)
		return io_u_accept;

	zbd_check_swd(f);

	pthread_mutex_lock(&zb->mutex);
	switch (io_u->ddir) {
	case DDIR_READ:
		if (td->runstate == TD_VERIFYING) {
			zb = zbd_replay_write_order(td, io_u, zb);
			goto accept;
		}
		/*
		 * Check that there is enough written data in the zone to do an
		 * I/O of at least min_bs B. If there isn't, find a new zone for
		 * the I/O.
		 */
		range = zb->cond != BLK_ZONE_COND_OFFLINE ?
			zb->wp - zb->start : 0;
		if (range < min_bs ||
		    ((!td_random(td)) && (io_u->offset + min_bs > zb->wp))) {
			pthread_mutex_unlock(&zb->mutex);
			zl = &f->zbd_info->zone_info[zbd_zone_idx(f,
						f->file_offset + f->io_size)];
			zb = zbd_find_zone(td, io_u, zb, zl);
			if (!zb) {
				dprint(FD_ZBD,
				       "%s: zbd_find_zone(%lld, %llu) failed\n",
				       f->file_name, io_u->offset,
				       io_u->buflen);
				goto eof;
			}
			/*
			 * zbd_find_zone() returned a zone with a range of at
			 * least min_bs.
			 */
			range = zb->wp - zb->start;
			assert(range >= min_bs);

			if (!td_random(td))
				io_u->offset = zb->start;
		}
		/*
		 * Make sure the I/O is within the zone valid data range while
		 * maximizing the I/O size and preserving randomness.
		 */
		if (range <= io_u->buflen)
			io_u->offset = zb->start;
		else if (td_random(td))
			io_u->offset = zb->start +
				((io_u->offset - orig_zb->start) %
				 (range - io_u->buflen)) / min_bs * min_bs;
		/*
		 * Make sure the I/O does not cross over the zone wp position.
		 */
		new_len = min((unsigned long long)io_u->buflen,
			      (unsigned long long)(zb->wp - io_u->offset));
		new_len = new_len / min_bs * min_bs;
		if (new_len < io_u->buflen) {
			io_u->buflen = new_len;
			dprint(FD_IO, "Changed length from %u into %llu\n",
			       orig_len, io_u->buflen);
		}
		assert(zb->start <= io_u->offset);
		assert(io_u->offset + io_u->buflen <= zb->wp);
		goto accept;
	case DDIR_WRITE:
		if (io_u->buflen > f->zbd_info->zone_size)
			goto eof;
		if (!zbd_open_zone(td, io_u, zone_idx_b)) {
			pthread_mutex_unlock(&zb->mutex);
			zb = zbd_convert_to_open_zone(td, io_u);
			if (!zb)
				goto eof;
			zone_idx_b = zb - f->zbd_info->zone_info;
		}
		/* Check whether the zone reset threshold has been exceeded */
		if (td->o.zrf.u.f) {
			if (f->zbd_info->sectors_with_data >=
			    f->io_size * td->o.zrt.u.f &&
			    zbd_dec_and_reset_write_cnt(td, f)) {
				zb->reset_zone = 1;
			}
		}
		/* Reset the zone pointer if necessary */
		if (zb->reset_zone || zbd_zone_full(f, zb, min_bs)) {
			assert(td->o.verify == VERIFY_NONE);
			/*
			 * Since previous write requests may have been submitted
			 * asynchronously and since we will submit the zone
			 * reset synchronously, wait until previously submitted
			 * write requests have completed before issuing a
			 * zone reset.
			 */
			io_u_quiesce(td);
			zb->reset_zone = 0;
			if (zbd_reset_zone(td, f, zb) < 0)
				goto eof;
		}
		/* Make writes occur at the write pointer */
		assert(!zbd_zone_full(f, zb, min_bs));
		io_u->offset = zb->wp;
		if (!is_valid_offset(f, io_u->offset)) {
			dprint(FD_ZBD, "Dropped request with offset %llu\n",
			       io_u->offset);
			goto eof;
		}
		/*
		 * Make sure that the buflen is a multiple of the minimal
		 * block size. Give up if shrinking would make the request too
		 * small.
		 */
		new_len = min((unsigned long long)io_u->buflen,
			      (zb + 1)->start - io_u->offset);
		new_len = new_len / min_bs * min_bs;
		if (new_len == io_u->buflen)
			goto accept;
		if (new_len >= min_bs) {
			io_u->buflen = new_len;
			dprint(FD_IO, "Changed length from %u into %llu\n",
			       orig_len, io_u->buflen);
			goto accept;
		}
		log_err("Zone remainder %lld smaller than minimum block size %d\n",
			((zb + 1)->start - io_u->offset),
			min_bs);
		goto eof;
	case DDIR_TRIM:
		/* fall-through */
	case DDIR_SYNC:
	case DDIR_DATASYNC:
	case DDIR_SYNC_FILE_RANGE:
	case DDIR_WAIT:
	case DDIR_LAST:
	case DDIR_INVAL:
		goto accept;
	}

	assert(false);

accept:
	assert(zb);
	assert(zb->cond != BLK_ZONE_COND_OFFLINE);
	assert(!io_u->post_submit);
	io_u->post_submit = zbd_post_submit;
	return io_u_accept;

eof:
	if (zb)
		pthread_mutex_unlock(&zb->mutex);
	return io_u_eof;
}

/* Return a string with ZBD statistics */
char *zbd_write_status(const struct thread_stat *ts)
{
	char *res;

	if (asprintf(&res, "; %llu zone resets", (unsigned long long) ts->nr_zone_resets) < 0)
		return NULL;
	return res;
}