X-Git-Url: https://git.kernel.dk/?p=fio.git;a=blobdiff_plain;f=zbd.c;h=8e99eb95dc6261101c48e6ea84ab5edf1601c316;hp=8dc3c39792931803b07d714ec222b1807c353102;hb=5ddf46d0b2dfe10b9a518db1f936c81e099b2646;hpb=c65057f93d1de62a48f98578e24716128ce77a75

diff --git a/zbd.c b/zbd.c
index 8dc3c397..8e99eb95 100644
--- a/zbd.c
+++ b/zbd.c
@@ -19,6 +19,7 @@
 #include "oslib/asprintf.h"
 #include "smalloc.h"
 #include "verify.h"
+#include "pshared.h"
 #include "zbd.h"
 
 /**
@@ -31,6 +32,17 @@ int zbd_get_zoned_model(struct thread_data *td, struct fio_file *f,
 {
 	int ret;
 
+	if (f->filetype == FIO_TYPE_PIPE) {
+		log_err("zonemode=zbd does not support pipes\n");
+		return -EINVAL;
+	}
+
+	/* If regular file, always emulate zones inside the file. */
+	if (f->filetype == FIO_TYPE_FILE) {
+		*model = ZBD_NONE;
+		return 0;
+	}
+
 	if (td->io_ops && td->io_ops->get_zoned_model)
 		ret = td->io_ops->get_zoned_model(td, f, model);
 	else
@@ -112,6 +124,34 @@ int zbd_reset_wp(struct thread_data *td, struct fio_file *f,
 	return ret;
 }
 
+/**
+ * zbd_get_max_open_zones - Get the maximum number of open zones
+ * @td: FIO thread data
+ * @f: FIO file for which to get max open zones
+ * @max_open_zones: Upon success, result will be stored here.
+ *
+ * A @max_open_zones value set to zero means no limit.
+ *
+ * Returns 0 upon success and a negative error code upon failure.
+ */
+int zbd_get_max_open_zones(struct thread_data *td, struct fio_file *f,
+			   unsigned int *max_open_zones)
+{
+	int ret;
+
+	if (td->io_ops && td->io_ops->get_max_open_zones)
+		ret = td->io_ops->get_max_open_zones(td, f, max_open_zones);
+	else
+		ret = blkzoned_get_max_open_zones(td, f, max_open_zones);
+	if (ret < 0) {
+		td_verror(td, errno, "get max open zones failed");
+		log_err("%s: get max open zones failed (%d).\n",
+			f->file_name, errno);
+	}
+
+	return ret;
+}
+
 /**
  * zbd_zone_idx - convert an offset into a zone number
  * @f: file pointer.
@@ -131,12 +171,21 @@ static uint32_t zbd_zone_idx(const struct fio_file *f, uint64_t offset)
 }
 
 /**
- * zbd_zone_swr - Test whether a zone requires sequential writes
+ * zbd_zone_end - Return zone end location
+ * @z: zone info pointer.
+ */
+static inline uint64_t zbd_zone_end(const struct fio_zone_info *z)
+{
+	return (z+1)->start;
+}
+
+/**
+ * zbd_zone_capacity_end - Return zone capacity limit end location
  * @z: zone info pointer.
  */
-static inline bool zbd_zone_swr(struct fio_zone_info *z)
+static inline uint64_t zbd_zone_capacity_end(const struct fio_zone_info *z)
 {
-	return z->type == ZBD_ZONE_TYPE_SWR;
+	return z->start + z->capacity;
 }
 
 /**
@@ -152,12 +201,21 @@ static bool zbd_zone_full(const struct fio_file *f, struct fio_zone_info *z,
 {
 	assert((required & 511) == 0);
 
-	return zbd_zone_swr(z) &&
-		z->wp + required > z->start + f->zbd_info->zone_size;
+	return z->has_wp &&
+		z->wp + required > zbd_zone_capacity_end(z);
 }
 
-static void zone_lock(struct thread_data *td, struct fio_zone_info *z)
+static void zone_lock(struct thread_data *td, const struct fio_file *f,
+		      struct fio_zone_info *z)
 {
+	struct zoned_block_device_info *zbd = f->zbd_info;
+	uint32_t nz = z - zbd->zone_info;
+
+	/* A thread should never lock zones outside its working area. */
+	assert(f->min_zone <= nz && nz < f->max_zone);
+
+	assert(z->has_wp);
+
 	/*
 	 * Lock the io_u target zone. The zone will be unlocked if io_u offset
 	 * is changed or when io_u completes and zbd_put_io() executed.
@@ -174,11 +232,26 @@ static void zone_lock(struct thread_data *td, struct fio_zone_info *z)
 	}
 }
 
+static inline void zone_unlock(struct fio_zone_info *z)
+{
+	int ret;
+
+	assert(z->has_wp);
+	ret = pthread_mutex_unlock(&z->mutex);
+	assert(!ret);
+}
+
 static bool is_valid_offset(const struct fio_file *f, uint64_t offset)
 {
 	return (uint64_t)(offset - f->file_offset) < f->io_size;
 }
 
+static inline struct fio_zone_info *get_zone(const struct fio_file *f,
+					     unsigned int zone_nr)
+{
+	return &f->zbd_info->zone_info[zone_nr];
+}
+
 /* Verify whether direct I/O is used for all host-managed zoned drives. */
 static bool zbd_using_direct_io(void)
 {
@@ -210,7 +283,7 @@ static bool zbd_is_seq_job(struct fio_file *f)
 	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 (zbd_zone_swr(&f->zbd_info->zone_info[zone_idx]))
+		if (get_zone(f, zone_idx)->has_wp)
 			return true;
 
 	return false;
@@ -251,9 +324,7 @@ static bool zbd_verify_sizes(void)
 				return false;
 			}
 
-			if (td->o.zone_skip &&
-			    (td->o.zone_skip < td->o.zone_size ||
-			     td->o.zone_skip % td->o.zone_size)) {
+			if (td->o.zone_skip % td->o.zone_size) {
 				log_err("%s: zoneskip %llu is not a multiple of the device zone size %llu.\n",
 					f->file_name, (unsigned long long) td->o.zone_skip,
 					(unsigned long long) td->o.zone_size);
@@ -261,9 +332,10 @@ static bool zbd_verify_sizes(void)
 			}
 
 			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;
+			z = get_zone(f, zone_idx);
+			if ((f->file_offset != z->start) &&
+			    (td->o.td_ddir != TD_DDIR_READ)) {
+				new_offset = zbd_zone_end(z);
 				if (new_offset >= f->file_offset + f->io_size) {
 					log_info("%s: io_size must be at least one zone\n",
 						 f->file_name);
@@ -276,9 +348,10 @@ static bool zbd_verify_sizes(void)
 				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];
+			z = get_zone(f, zone_idx);
 			new_end = z->start;
-			if (f->file_offset + f->io_size != new_end) {
+			if ((td->o.td_ddir != TD_DDIR_READ) &&
+			    (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);
@@ -299,20 +372,21 @@ 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) {
+			uint64_t zone_size;
+
 			if (!f->zbd_info)
 				continue;
 			zone_size = f->zbd_info->zone_size;
-			for (k = 0; k < ARRAY_SIZE(td->o.bs); k++) {
+			for (k = 0; k < FIO_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",
+					log_info("%s: block size %llu is not a divisor of the zone size %llu\n",
 						 f->file_name, td->o.bs[k],
-						 zone_size);
+						 (unsigned long long)zone_size);
 					return false;
 				}
 			}
@@ -341,12 +415,12 @@ 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 = td->o.zone_size;
+	uint64_t zone_capacity = td->o.zone_capacity;
 	struct zoned_block_device_info *zbd_info = NULL;
-	pthread_mutexattr_t attr;
 	int i;
 
 	if (zone_size == 0) {
-		log_err("%s: Specifying the zone size is mandatory for regular block devices with --zonemode=zbd\n\n",
+		log_err("%s: Specifying the zone size is mandatory for regular file/block device with --zonemode=zbd\n\n",
 			f->file_name);
 		return 1;
 	}
@@ -357,24 +431,40 @@ static int init_zone_info(struct thread_data *td, struct fio_file *f)
 		return 1;
 	}
 
+	if (zone_capacity == 0)
+		zone_capacity = zone_size;
+
+	if (zone_capacity > zone_size) {
+		log_err("%s: job parameter zonecapacity %llu is larger than zone size %llu\n",
+			f->file_name, (unsigned long long) td->o.zone_capacity,
+			(unsigned long long) td->o.zone_size);
+		return 1;
+	}
+
+	if (f->real_file_size < zone_size) {
+		log_err("%s: file/device size %"PRIu64" is smaller than zone size %"PRIu64"\n",
+			f->file_name, f->real_file_size, zone_size);
+		return -EINVAL;
+	}
+
 	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);
+	mutex_init_pshared(&zbd_info->mutex);
 	zbd_info->refcount = 1;
 	p = &zbd_info->zone_info[0];
 	for (i = 0; i < nr_zones; i++, p++) {
-		pthread_mutex_init(&p->mutex, &attr);
+		mutex_init_pshared_with_type(&p->mutex,
+					     PTHREAD_MUTEX_RECURSIVE);
 		p->start = i * zone_size;
-		p->wp = p->start + zone_size;
+		p->wp = p->start;
 		p->type = ZBD_ZONE_TYPE_SWR;
 		p->cond = ZBD_ZONE_COND_EMPTY;
+		p->capacity = zone_capacity;
+		p->has_wp = 1;
 	}
 	/* a sentinel */
 	p->start = nr_zones * zone_size;
@@ -384,7 +474,6 @@ static int init_zone_info(struct thread_data *td, struct fio_file *f)
 	f->zbd_info->zone_size_log2 = is_power_of_2(zone_size) ?
 		ilog2(zone_size) : 0;
 	f->zbd_info->nr_zones = nr_zones;
-	pthread_mutexattr_destroy(&attr);
 	return 0;
 }
 
@@ -404,12 +493,7 @@ static int parse_zone_info(struct thread_data *td, struct fio_file *f)
 	struct fio_zone_info *p;
 	uint64_t zone_size, offset;
 	struct zoned_block_device_info *zbd_info = NULL;
-	pthread_mutexattr_t attr;
-	int i, j, ret = 0;
-
-	pthread_mutexattr_init(&attr);
-	pthread_mutexattr_settype(&attr, PTHREAD_MUTEX_RECURSIVE);
-	pthread_mutexattr_setpshared(&attr, true);
+	int i, j, ret = -ENOMEM;
 
 	zones = calloc(ZBD_REPORT_MAX_ZONES, sizeof(struct zbd_zone));
 	if (!zones)
@@ -441,21 +525,22 @@ static int parse_zone_info(struct thread_data *td, struct fio_file *f)
 
 	zbd_info = scalloc(1, sizeof(*zbd_info) +
 			   (nr_zones + 1) * sizeof(zbd_info->zone_info[0]));
-	ret = -ENOMEM;
 	if (!zbd_info)
 		goto out;
-	pthread_mutex_init(&zbd_info->mutex, &attr);
+	mutex_init_pshared(&zbd_info->mutex);
 	zbd_info->refcount = 1;
 	p = &zbd_info->zone_info[0];
 	for (offset = 0, j = 0; j < nr_zones;) {
 		z = &zones[0];
 		for (i = 0; i < nrz; i++, j++, z++, p++) {
-			pthread_mutex_init(&p->mutex, &attr);
+			mutex_init_pshared_with_type(&p->mutex,
+						     PTHREAD_MUTEX_RECURSIVE);
 			p->start = z->start;
+			p->capacity = z->capacity;
 			switch (z->cond) {
 			case ZBD_ZONE_COND_NOT_WP:
 			case ZBD_ZONE_COND_FULL:
-				p->wp = p->start + zone_size;
+				p->wp = p->start + p->capacity;
 				break;
 			default:
 				assert(z->start <= z->wp);
@@ -463,8 +548,17 @@ static int parse_zone_info(struct thread_data *td, struct fio_file *f)
 				p->wp = z->wp;
 				break;
 			}
+
+			switch (z->type) {
+			case ZBD_ZONE_TYPE_SWR:
+				p->has_wp = 1;
+				break;
+			default:
+				p->has_wp = 0;
+			}
 			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);
@@ -476,8 +570,9 @@ static int parse_zone_info(struct thread_data *td, struct fio_file *f)
 		offset = z->start + z->len;
 		if (j >= nr_zones)
 			break;
-		nrz = zbd_report_zones(td, f, offset,
-					    zones, ZBD_REPORT_MAX_ZONES);
+		nrz = zbd_report_zones(td, f, offset, zones,
+				       min((uint32_t)(nr_zones - j),
+					   ZBD_REPORT_MAX_ZONES));
 		if (nrz < 0) {
 			ret = nrz;
 			log_info("fio: report zones (offset %llu) failed for %s (%d).\n",
@@ -501,10 +596,54 @@ static int parse_zone_info(struct thread_data *td, struct fio_file *f)
 out:
 	sfree(zbd_info);
 	free(zones);
-	pthread_mutexattr_destroy(&attr);
 	return ret;
 }
 
+static int zbd_set_max_open_zones(struct thread_data *td, struct fio_file *f)
+{
+	struct zoned_block_device_info *zbd = f->zbd_info;
+	unsigned int max_open_zones;
+	int ret;
+
+	if (zbd->model != ZBD_HOST_MANAGED || td->o.ignore_zone_limits) {
+		/* Only host-managed devices have a max open limit */
+		zbd->max_open_zones = td->o.max_open_zones;
+		goto out;
+	}
+
+	/* If host-managed, get the max open limit */
+	ret = zbd_get_max_open_zones(td, f, &max_open_zones);
+	if (ret)
+		return ret;
+
+	if (!max_open_zones) {
+		/* No device limit */
+		zbd->max_open_zones = td->o.max_open_zones;
+	} else if (!td->o.max_open_zones) {
+		/* No user limit. Set limit to device limit */
+		zbd->max_open_zones = max_open_zones;
+	} else if (td->o.max_open_zones <= max_open_zones) {
+		/* Both user limit and dev limit. User limit not too large */
+		zbd->max_open_zones = td->o.max_open_zones;
+	} else {
+		/* Both user limit and dev limit. User limit too large */
+		td_verror(td, EINVAL,
+			  "Specified --max_open_zones is too large");
+		log_err("Specified --max_open_zones (%d) is larger than max (%u)\n",
+			td->o.max_open_zones, max_open_zones);
+		return -EINVAL;
+	}
+
+out:
+	/* Ensure that the limit is not larger than FIO's internal limit */
+	zbd->max_open_zones = min_not_zero(zbd->max_open_zones,
+					   (uint32_t) ZBD_MAX_OPEN_ZONES);
+	dprint(FD_ZBD, "%s: using max open zones limit: %"PRIu32"\n",
+	       f->file_name, zbd->max_open_zones);
+
+	return 0;
+}
+
 /*
  * Allocate zone information and store it into f->zbd_info if zonemode=zbd.
  *
@@ -522,14 +661,16 @@ static int zbd_create_zone_info(struct thread_data *td, struct fio_file *f)
 		return ret;
 
 	switch (zbd_model) {
-	case ZBD_IGNORE:
-		return 0;
 	case ZBD_HOST_AWARE:
 	case ZBD_HOST_MANAGED:
 		ret = parse_zone_info(td, f);
+		if (ret)
+			return ret;
 		break;
 	case ZBD_NONE:
 		ret = init_zone_info(td, f);
+		if (ret)
+			return ret;
 		break;
 	default:
 		td_verror(td, EINVAL, "Unsupported zoned model");
@@ -537,17 +678,23 @@ static int zbd_create_zone_info(struct thread_data *td, struct fio_file *f)
 		return -EINVAL;
 	}
 
-	if (ret == 0)
-		f->zbd_info->model = zbd_model;
-	return ret;
+	assert(f->zbd_info);
+	f->zbd_info->model = zbd_model;
+
+	ret = zbd_set_max_open_zones(td, f);
+	if (ret) {
+		zbd_free_zone_info(f);
+		return ret;
+	}
+
+	return 0;
 }
 
 void zbd_free_zone_info(struct fio_file *f)
 {
 	uint32_t refcount;
 
-	if (!f->zbd_info)
-		return;
+	assert(f->zbd_info);
 
 	pthread_mutex_lock(&f->zbd_info->mutex);
 	refcount = --f->zbd_info->refcount;
@@ -592,7 +739,12 @@ static int zbd_init_zone_info(struct thread_data *td, struct fio_file *file)
 	return ret;
 }
 
-int zbd_init(struct thread_data *td)
+static bool zbd_open_zone(struct thread_data *td, const struct fio_file *f,
+			  uint32_t zone_idx);
+static int zbd_reset_zone(struct thread_data *td, struct fio_file *f,
+			  struct fio_zone_info *z);
+
+int zbd_init_files(struct thread_data *td)
 {
 	struct fio_file *f;
 	int i;
@@ -601,6 +753,44 @@ int zbd_init(struct thread_data *td)
 		if (zbd_init_zone_info(td, f))
 			return 1;
 	}
+	return 0;
+}
+
+void zbd_recalc_options_with_zone_granularity(struct thread_data *td)
+{
+	struct fio_file *f;
+	int i;
+
+	for_each_file(td, f, i) {
+		struct zoned_block_device_info *zbd = f->zbd_info;
+		// zonemode=strided doesn't get per-file zone size.
+		uint64_t zone_size = zbd ? zbd->zone_size : td->o.zone_size;
+
+		if (zone_size == 0)
+			continue;
+
+		if (td->o.size_nz > 0) {
+			td->o.size = td->o.size_nz * zone_size;
+		}
+		if (td->o.io_size_nz > 0) {
+			td->o.io_size = td->o.io_size_nz * zone_size;
+		}
+		if (td->o.start_offset_nz > 0) {
+			td->o.start_offset = td->o.start_offset_nz * zone_size;
+		}
+		if (td->o.offset_increment_nz > 0) {
+			td->o.offset_increment = td->o.offset_increment_nz * zone_size;
+		}
+		if (td->o.zone_skip_nz > 0) {
+			td->o.zone_skip = td->o.zone_skip_nz * zone_size;
+		}
+	}
+}
+
+int zbd_setup_files(struct thread_data *td)
+{
+	struct fio_file *f;
+	int i;
 
 	if (!zbd_using_direct_io()) {
 		log_err("Using direct I/O is mandatory for writing to ZBD drives\n\n");
@@ -613,27 +803,87 @@ int zbd_init(struct thread_data *td)
 	if (!zbd_verify_bs())
 		return 1;
 
+	for_each_file(td, f, i) {
+		struct zoned_block_device_info *zbd = f->zbd_info;
+		struct fio_zone_info *z;
+		int zi;
+
+		assert(zbd);
+
+		f->min_zone = zbd_zone_idx(f, f->file_offset);
+		f->max_zone = zbd_zone_idx(f, f->file_offset + f->io_size);
+
+		/*
+		 * When all zones in the I/O range are conventional, io_size
+		 * can be smaller than zone size, making min_zone the same
+		 * as max_zone. This is why the assert below needs to be made
+		 * conditional.
+		 */
+		if (zbd_is_seq_job(f))
+			assert(f->min_zone < f->max_zone);
+
+		if (td->o.max_open_zones > 0 &&
+		    zbd->max_open_zones != td->o.max_open_zones) {
+			log_err("Different 'max_open_zones' values\n");
+			return 1;
+		}
+		if (zbd->max_open_zones > ZBD_MAX_OPEN_ZONES) {
+			log_err("'max_open_zones' value is limited by %u\n", ZBD_MAX_OPEN_ZONES);
+			return 1;
+		}
+
+		for (zi = f->min_zone; zi < f->max_zone; zi++) {
+			z = &zbd->zone_info[zi];
+			if (z->cond != ZBD_ZONE_COND_IMP_OPEN &&
+			    z->cond != ZBD_ZONE_COND_EXP_OPEN)
+				continue;
+			if (zbd_open_zone(td, f, zi))
+				continue;
+			/*
+			 * If the number of open zones exceeds specified limits,
+			 * reset all extra open zones.
+			 */
+			if (zbd_reset_zone(td, f, z) < 0) {
+				log_err("Failed to reest zone %d\n", zi);
+				return 1;
+			}
+		}
+	}
+
 	return 0;
 }
 
+static inline unsigned int zbd_zone_nr(const struct fio_file *f,
+				       struct fio_zone_info *zone)
+{
+	return zone - f->zbd_info->zone_info;
+}
+
 /**
- * zbd_reset_range - reset zones for a range of sectors
+ * zbd_reset_zone - reset the write pointer of a single zone
  * @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
+ * @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.
+ *
+ * The caller must hold z->mutex.
  */
-static int zbd_reset_range(struct thread_data *td, struct fio_file *f,
-			   uint64_t offset, uint64_t length)
+static int zbd_reset_zone(struct thread_data *td, struct fio_file *f,
+			  struct fio_zone_info *z)
 {
-	uint32_t zone_idx_b, zone_idx_e;
-	struct fio_zone_info *zb, *ze, *z;
+	uint64_t offset = z->start;
+	uint64_t length = (z+1)->start - offset;
+	uint64_t data_in_zone = z->wp - z->start;
 	int ret = 0;
 
+	if (!data_in_zone)
+		return 0;
+
 	assert(is_valid_offset(f, offset + length - 1));
 
+	dprint(FD_ZBD, "%s: resetting wp of zone %u.\n", f->file_name,
+		zbd_zone_nr(f, z));
 	switch (f->zbd_info->model) {
 	case ZBD_HOST_AWARE:
 	case ZBD_HOST_MANAGED:
@@ -645,62 +895,39 @@ static int zbd_reset_range(struct thread_data *td, struct fio_file *f,
 		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);
-	}
+	pthread_mutex_lock(&f->zbd_info->mutex);
+	f->zbd_info->sectors_with_data -= data_in_zone;
+	f->zbd_info->wp_sectors_with_data -= data_in_zone;
+	pthread_mutex_unlock(&f->zbd_info->mutex);
+	z->wp = z->start;
+	z->verify_block = 0;
 
-	td->ts.nr_zone_resets += ze - zb;
+	td->ts.nr_zone_resets++;
 
 	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, 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);
-}
-
 /* 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)
+			   unsigned int zone_idx)
 {
-	uint32_t zone_idx;
+	uint32_t open_zone_idx = 0;
 
-	assert(open_zone_idx < f->zbd_info->num_open_zones);
-	zone_idx = f->zbd_info->open_zones[open_zone_idx];
+	for (; open_zone_idx < f->zbd_info->num_open_zones; open_zone_idx++) {
+		if (f->zbd_info->open_zones[open_zone_idx] == zone_idx)
+			break;
+	}
+	if (open_zone_idx == f->zbd_info->num_open_zones)
+		return;
+
+	dprint(FD_ZBD, "%s: closing zone %d\n", f->file_name, zone_idx);
 	memmove(f->zbd_info->open_zones + open_zone_idx,
 		f->zbd_info->open_zones + open_zone_idx + 1,
 		(ZBD_MAX_OPEN_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;
+	td->num_open_zones--;
+	get_zone(f, zone_idx)->open = 0;
 }
 
 /*
@@ -709,51 +936,35 @@ static void zbd_close_zone(struct thread_data *td, const struct fio_file *f,
  * @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 *const ze)
 {
 	struct fio_zone_info *z;
 	const uint32_t min_bs = td->o.min_bs[DDIR_WRITE];
-	bool reset_wp;
 	int res = 0;
 
 	assert(min_bs);
 
 	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));
+		zbd_zone_nr(f, zb), zbd_zone_nr(f, ze));
 	for (z = zb; z < ze; z++) {
-		uint32_t nz = z - f->zbd_info->zone_info;
+		uint32_t nz = zbd_zone_nr(f, z);
 
-		if (!zbd_zone_swr(z))
+		if (!z->has_wp)
 			continue;
-		zone_lock(td, z);
-		if (all_zones) {
-			unsigned int i;
-
-			pthread_mutex_lock(&f->zbd_info->mutex);
-			for (i = 0; i < f->zbd_info->num_open_zones; i++) {
-				if (f->zbd_info->open_zones[i] == nz)
-					zbd_close_zone(td, f, i);
-			}
-			pthread_mutex_unlock(&f->zbd_info->mutex);
-
-			reset_wp = z->wp != z->start;
-		} else {
-			reset_wp = (td->o.td_ddir & TD_DDIR_WRITE) &&
-					z->wp % min_bs != 0;
-		}
-		if (reset_wp) {
+		zone_lock(td, f, z);
+		pthread_mutex_lock(&f->zbd_info->mutex);
+		zbd_close_zone(td, f, nz);
+		pthread_mutex_unlock(&f->zbd_info->mutex);
+		if (z->wp != z->start) {
 			dprint(FD_ZBD, "%s: resetting zone %u\n",
-			       f->file_name,
-			       zbd_zone_nr(f->zbd_info, z));
+			       f->file_name, zbd_zone_nr(f, z));
 			if (zbd_reset_zone(td, f, z) < 0)
 				res = 1;
 		}
-		pthread_mutex_unlock(&z->mutex);
+		zone_unlock(z);
 	}
 
 	return res;
@@ -763,14 +974,20 @@ static int zbd_reset_zones(struct thread_data *td, struct fio_file *f,
  * 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)
+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;
+}
+
+static void zbd_reset_write_cnt(const struct thread_data *td,
+				const struct fio_file *f)
+{
+	pthread_mutex_lock(&f->zbd_info->mutex);
+	_zbd_reset_write_cnt(td, f);
 	pthread_mutex_unlock(&f->zbd_info->mutex);
 }
 
@@ -784,7 +1001,7 @@ static bool zbd_dec_and_reset_write_cnt(const struct thread_data *td,
 	if (f->zbd_info->write_cnt)
 		write_cnt = --f->zbd_info->write_cnt;
 	if (write_cnt == 0)
-		zbd_reset_write_cnt(td, f);
+		_zbd_reset_write_cnt(td, f);
 	pthread_mutex_unlock(&f->zbd_info->mutex);
 
 	return write_cnt == 0;
@@ -796,30 +1013,37 @@ enum swd_action {
 };
 
 /* 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)
+static uint64_t zbd_process_swd(struct thread_data *td,
+				const struct fio_file *f, enum swd_action a)
 {
 	struct fio_zone_info *zb, *ze, *z;
 	uint64_t swd = 0;
+	uint64_t wp_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)];
+	zb = get_zone(f, f->min_zone);
+	ze = get_zone(f, f->max_zone);
 	for (z = zb; z < ze; z++) {
-		pthread_mutex_lock(&z->mutex);
+		if (z->has_wp) {
+			zone_lock(td, f, z);
+			wp_swd += z->wp - z->start;
+		}
 		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);
+		assert(f->zbd_info->wp_sectors_with_data == wp_swd);
 		break;
 	case SET_SWD:
 		f->zbd_info->sectors_with_data = swd;
+		f->zbd_info->wp_sectors_with_data = wp_swd;
 		break;
 	}
 	pthread_mutex_unlock(&f->zbd_info->mutex);
 	for (z = zb; z < ze; z++)
-		pthread_mutex_unlock(&z->mutex);
+		if (z->has_wp)
+			zone_unlock(z);
 
 	return swd;
 }
@@ -830,47 +1054,35 @@ static uint64_t zbd_process_swd(const struct fio_file *f, enum swd_action a)
  */
 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)
+/* Check whether the values of zbd_info.*sectors_with_data are correct. */
+static void zbd_check_swd(struct thread_data *td, const struct fio_file *f)
 {
 	if (!enable_check_swd)
 		return;
 
-	zbd_process_swd(f, CHECK_SWD);
+	zbd_process_swd(td, f, CHECK_SWD);
 }
 
-static void zbd_init_swd(struct fio_file *f)
+void zbd_file_reset(struct thread_data *td, struct fio_file *f)
 {
+	struct fio_zone_info *zb, *ze;
 	uint64_t swd;
 
-	if (!enable_check_swd)
+	if (!f->zbd_info || !td_write(td))
 		return;
 
-	swd = zbd_process_swd(f, SET_SWD);
+	zb = get_zone(f, f->min_zone);
+	ze = get_zone(f, f->max_zone);
+	swd = zbd_process_swd(td, 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);
+	if (td->o.verify != VERIFY_NONE && td->runstate != TD_VERIFYING)
+		zbd_reset_zones(td, f, zb, ze);
 	zbd_reset_write_cnt(td, f);
 }
 
@@ -881,8 +1093,9 @@ static bool is_zone_open(const struct thread_data *td, const struct fio_file *f,
 	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);
+	assert(td->o.job_max_open_zones == 0 || td->num_open_zones <= td->o.job_max_open_zones);
+	assert(td->o.job_max_open_zones <= zbdi->max_open_zones);
+	assert(zbdi->num_open_zones <= zbdi->max_open_zones);
 
 	for (i = 0; i < zbdi->num_open_zones; i++)
 		if (zbdi->open_zones[i] == zone_idx)
@@ -897,12 +1110,11 @@ static bool is_zone_open(const struct thread_data *td, const struct fio_file *f,
  * 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,
+static bool zbd_open_zone(struct thread_data *td, const struct fio_file *f,
 			  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];
+	struct fio_zone_info *z = get_zone(f, zone_idx);
 	bool res = true;
 
 	if (z->cond == ZBD_ZONE_COND_OFFLINE)
@@ -915,18 +1127,26 @@ static bool zbd_open_zone(struct thread_data *td, const struct io_u *io_u,
 	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))
+	if (is_zone_open(td, f, zone_idx)) {
+		/*
+		 * If the zone is already open and going to be full by writes
+		 * in-flight, handle it as a full zone instead of an open zone.
+		 */
+		if (z->wp >= zbd_zone_capacity_end(z))
+			res = false;
 		goto out;
+	}
 	res = false;
-	if (f->zbd_info->num_open_zones >= td->o.max_open_zones)
+	/* Zero means no limit */
+	if (td->o.job_max_open_zones > 0 &&
+	    td->num_open_zones >= td->o.job_max_open_zones)
+		goto out;
+	if (f->zbd_info->num_open_zones >= f->zbd_info->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;
+	td->num_open_zones++;
 	z->open = 1;
 	res = true;
 
@@ -945,7 +1165,8 @@ static uint32_t pick_random_zone_idx(const struct fio_file *f,
 /*
  * 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
+ * necessary. The open zone is searched across sequential zones.
+ * 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.
  */
@@ -953,15 +1174,16 @@ static 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_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;
+	bool wait_zone_close;
 
 	assert(is_valid_offset(f, io_u->offset));
 
-	if (td->o.max_open_zones) {
+	if (td->o.max_open_zones || td->o.job_max_open_zones) {
 		/*
 		 * This statement accesses f->zbd_info->open_zones[] on purpose
 		 * without locking.
@@ -970,6 +1192,10 @@ static struct fio_zone_info *zbd_convert_to_open_zone(struct thread_data *td,
 	} else {
 		zone_idx = zbd_zone_idx(f, io_u->offset);
 	}
+	if (zone_idx < f->min_zone)
+		zone_idx = f->min_zone;
+	else if (zone_idx >= f->max_zone)
+		zone_idx = f->max_zone - 1;
 	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);
 
@@ -982,18 +1208,19 @@ static struct fio_zone_info *zbd_convert_to_open_zone(struct thread_data *td,
 	for (;;) {
 		uint32_t tmp_idx;
 
-		z = &f->zbd_info->zone_info[zone_idx];
-
-		zone_lock(td, z);
+		z = get_zone(f, zone_idx);
+		if (z->has_wp)
+			zone_lock(td, f, z);
 		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;
+		if (z->has_wp) {
+			if (z->cond != ZBD_ZONE_COND_OFFLINE &&
+			    td->o.max_open_zones == 0 && td->o.job_max_open_zones == 0)
+				goto examine_zone;
+			if (f->zbd_info->num_open_zones == 0) {
+				dprint(FD_ZBD, "%s(%s): no zones are open\n",
+				       __func__, f->file_name);
+				goto open_other_zone;
+			}
 		}
 
 		/*
@@ -1002,7 +1229,8 @@ static struct fio_zone_info *zbd_convert_to_open_zone(struct thread_data *td,
 		 * Ignore zones which don't belong to thread's offset/size area.
 		 */
 		open_zone_idx = pick_random_zone_idx(f, io_u);
-		assert(open_zone_idx < f->zbd_info->num_open_zones);
+		assert(!open_zone_idx ||
+		       open_zone_idx < f->zbd_info->num_open_zones);
 		tmp_idx = open_zone_idx;
 		for (i = 0; i < f->zbd_info->num_open_zones; i++) {
 			uint32_t tmpz;
@@ -1010,8 +1238,7 @@ static struct fio_zone_info *zbd_convert_to_open_zone(struct thread_data *td,
 			if (tmp_idx >= f->zbd_info->num_open_zones)
 				tmp_idx = 0;
 			tmpz = f->zbd_info->open_zones[tmp_idx];
-
-			if (is_valid_offset(f, f->zbd_info->zone_info[tmpz].start)) {
+			if (f->min_zone <= tmpz && tmpz < f->max_zone) {
 				open_zone_idx = tmp_idx;
 				goto found_candidate_zone;
 			}
@@ -1022,7 +1249,8 @@ static struct fio_zone_info *zbd_convert_to_open_zone(struct thread_data *td,
 		dprint(FD_ZBD, "%s(%s): no candidate zone\n",
 			__func__, f->file_name);
 		pthread_mutex_unlock(&f->zbd_info->mutex);
-		pthread_mutex_unlock(&z->mutex);
+		if (z->has_wp)
+			zone_unlock(z);
 		return NULL;
 
 found_candidate_zone:
@@ -1031,39 +1259,59 @@ found_candidate_zone:
 			break;
 		zone_idx = new_zone_idx;
 		pthread_mutex_unlock(&f->zbd_info->mutex);
-		pthread_mutex_unlock(&z->mutex);
+		if (z->has_wp)
+			zone_unlock(z);
 	}
 
 	/* Both z->mutex and f->zbd_info->mutex are held. */
 
 examine_zone:
-	if (z->wp + min_bs <= (z+1)->start) {
+	if (z->wp + min_bs <= zbd_zone_capacity_end(z)) {
 		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);
+
+open_other_zone:
+	/* Check if number of open zones reaches one of limits. */
+	wait_zone_close =
+		f->zbd_info->num_open_zones == f->max_zone - f->min_zone ||
+		(td->o.max_open_zones &&
+		 f->zbd_info->num_open_zones == td->o.max_open_zones) ||
+		(td->o.job_max_open_zones &&
+		 td->num_open_zones == td->o.job_max_open_zones);
+
 	pthread_mutex_unlock(&f->zbd_info->mutex);
 
 	/* Only z->mutex is held. */
 
+	/*
+	 * When number of open zones reaches to one of limits, wait for
+	 * zone close before opening a new zone.
+	 */
+	if (wait_zone_close) {
+		dprint(FD_ZBD, "%s(%s): quiesce to allow open zones to close\n",
+		       __func__, f->file_name);
+		io_u_quiesce(td);
+	}
+
 	/* 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);
+		if (z->has_wp)
+			zone_unlock(z);
 		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];
+			zone_idx = f->min_zone;
+			z = get_zone(f, zone_idx);
 		}
 		assert(is_valid_offset(f, z->start));
-		zone_lock(td, z);
+		if (!z->has_wp)
+			continue;
+		zone_lock(td, f, z);
 		if (z->open)
 			continue;
-		if (zbd_open_zone(td, io_u, zone_idx))
+		if (zbd_open_zone(td, f, zone_idx))
 			goto out;
 	}
 
@@ -1073,18 +1321,20 @@ examine_zone:
 	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];
+		if (zone_idx < f->min_zone || zone_idx >= f->max_zone)
+			continue;
 		pthread_mutex_unlock(&f->zbd_info->mutex);
-		pthread_mutex_unlock(&z->mutex);
+		zone_unlock(z);
 
-		z = &f->zbd_info->zone_info[zone_idx];
+		z = get_zone(f, zone_idx);
 
-		zone_lock(td, z);
-		if (z->wp + min_bs <= (z+1)->start)
+		zone_lock(td, f, z);
+		if (z->wp + min_bs <= zbd_zone_capacity_end(z))
 			goto out;
 		pthread_mutex_lock(&f->zbd_info->mutex);
 	}
 	pthread_mutex_unlock(&f->zbd_info->mutex);
-	pthread_mutex_unlock(&z->mutex);
+	zone_unlock(z);
 	dprint(FD_ZBD, "%s(%s): did not open another zone\n", __func__,
 	       f->file_name);
 	return NULL;
@@ -1093,6 +1343,8 @@ out:
 	dprint(FD_ZBD, "%s(%s): returning zone %d\n", __func__, f->file_name,
 	       zone_idx);
 	io_u->offset = z->start;
+	assert(z->has_wp);
+	assert(z->cond != ZBD_ZONE_COND_OFFLINE);
 	return z;
 }
 
@@ -1104,36 +1356,48 @@ static struct fio_zone_info *zbd_replay_write_order(struct thread_data *td,
 	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);
+	if (!zbd_open_zone(td, f, zbd_zone_nr(f, z))) {
+		zone_unlock(z);
 		z = zbd_convert_to_open_zone(td, io_u);
 		assert(z);
 	}
 
-	if (z->verify_block * min_bs >= f->zbd_info->zone_size)
+	if (z->verify_block * min_bs >= z->capacity) {
 		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;
+			min_bs, (unsigned long long)z->capacity);
+		/*
+		 * If the assertion below fails during a test run, adding
+		 * "--experimental_verify=1" to the command line may help.
+		 */
+		assert(false);
+	}
+	io_u->offset = z->start + z->verify_block * min_bs;
+	if (io_u->offset + io_u->buflen >= zbd_zone_capacity_end(z)) {
+		log_err("%s: %llu + %llu >= %llu\n", f->file_name, io_u->offset,
+			io_u->buflen, (unsigned long long) zbd_zone_capacity_end(z));
+		assert(false);
+	}
+	z->verify_block += io_u->buflen / 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.
+ * Find another zone for which @io_u fits in the readable data in the zone.
+ * Search in zones @zb + 1 .. @zl. For random workload, also search in zones
+ * @zb - 1 .. @zf.
  *
- * Either returns NULL or returns a zone pointer and holds the mutex for that
- * zone.
+ * Either returns NULL or returns a zone pointer. When the zone has write
+ * pointer, hold the mutex for the 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_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)];
+	const struct fio_zone_info *const zf = get_zone(f, f->min_zone);
 
 	/*
 	 * Skip to the next non-empty zone in case of sequential I/O and to
@@ -1141,19 +1405,23 @@ zbd_find_zone(struct thread_data *td, struct io_u *io_u,
 	 */
 	for (z1 = zb + 1, z2 = zb - 1; z1 < zl || z2 >= zf; z1++, z2--) {
 		if (z1 < zl && z1->cond != ZBD_ZONE_COND_OFFLINE) {
-			pthread_mutex_lock(&z1->mutex);
+			if (z1->has_wp)
+				zone_lock(td, f, z1);
 			if (z1->start + min_bs <= z1->wp)
 				return z1;
-			pthread_mutex_unlock(&z1->mutex);
+			if (z1->has_wp)
+				zone_unlock(z1);
 		} else if (!td_random(td)) {
 			break;
 		}
 		if (td_random(td) && z2 >= zf &&
 		    z2->cond != ZBD_ZONE_COND_OFFLINE) {
-			pthread_mutex_lock(&z2->mutex);
+			if (z2->has_wp)
+				zone_lock(td, f, z2);
 			if (z2->start + min_bs <= z2->wp)
 				return z2;
-			pthread_mutex_unlock(&z2->mutex);
+			if (z2->has_wp)
+				zone_unlock(z2);
 		}
 	}
 	dprint(FD_ZBD, "%s: adjusting random read offset failed\n",
@@ -1161,6 +1429,28 @@ zbd_find_zone(struct thread_data *td, struct io_u *io_u,
 	return NULL;
 }
 
+/**
+ * zbd_end_zone_io - update zone status at command completion
+ * @io_u: I/O unit
+ * @z: zone info pointer
+ *
+ * If the write command made the zone full, close it.
+ *
+ * The caller must hold z->mutex.
+ */
+static void zbd_end_zone_io(struct thread_data *td, const struct io_u *io_u,
+			    struct fio_zone_info *z)
+{
+	const struct fio_file *f = io_u->file;
+
+	if (io_u->ddir == DDIR_WRITE &&
+	    io_u->offset + io_u->buflen >= zbd_zone_capacity_end(z)) {
+		pthread_mutex_lock(&f->zbd_info->mutex);
+		zbd_close_zone(td, f, zbd_zone_nr(f, z));
+		pthread_mutex_unlock(&f->zbd_info->mutex);
+	}
+}
+
 /**
  * zbd_queue_io - update the write pointer of a sequential zone
  * @io_u: I/O unit
@@ -1170,7 +1460,8 @@ zbd_find_zone(struct thread_data *td, struct io_u *io_u,
  * For write and trim operations, update the write pointer of the I/O unit
  * target zone.
  */
-static void zbd_queue_io(struct io_u *io_u, int q, bool success)
+static void zbd_queue_io(struct thread_data *td, struct io_u *io_u, int q,
+			 bool success)
 {
 	const struct fio_file *f = io_u->file;
 	struct zoned_block_device_info *zbd_info = f->zbd_info;
@@ -1178,15 +1469,13 @@ static void zbd_queue_io(struct io_u *io_u, int q, bool success)
 	uint32_t zone_idx;
 	uint64_t zone_end;
 
-	if (!zbd_info)
-		return;
+	assert(zbd_info);
 
 	zone_idx = zbd_zone_idx(f, io_u->offset);
 	assert(zone_idx < zbd_info->nr_zones);
-	z = &zbd_info->zone_info[zone_idx];
+	z = get_zone(f, zone_idx);
 
-	if (!zbd_zone_swr(z))
-		return;
+	assert(z->has_wp);
 
 	if (!success)
 		goto unlock;
@@ -1198,14 +1487,16 @@ static void zbd_queue_io(struct io_u *io_u, int q, bool success)
 	switch (io_u->ddir) {
 	case DDIR_WRITE:
 		zone_end = min((uint64_t)(io_u->offset + io_u->buflen),
-			       (z + 1)->start);
+			       zbd_zone_capacity_end(z));
 		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)
+		if (z->wp <= zone_end) {
 			zbd_info->sectors_with_data += zone_end - z->wp;
+			zbd_info->wp_sectors_with_data += zone_end - z->wp;
+		}
 		pthread_mutex_unlock(&zbd_info->mutex);
 		z->wp = zone_end;
 		break;
@@ -1216,10 +1507,13 @@ static void zbd_queue_io(struct io_u *io_u, int q, bool success)
 		break;
 	}
 
+	if (q == FIO_Q_COMPLETED && !io_u->error)
+		zbd_end_zone_io(td, io_u, z);
+
 unlock:
 	if (!success || q != FIO_Q_QUEUED) {
 		/* BUSY or COMPLETED: unlock the zone */
-		pthread_mutex_unlock(&z->mutex);
+		zone_unlock(z);
 		io_u->zbd_put_io = NULL;
 	}
 }
@@ -1228,31 +1522,29 @@ unlock:
  * zbd_put_io - Unlock an I/O unit target zone lock
  * @io_u: I/O unit
  */
-static void zbd_put_io(const struct io_u *io_u)
+static void zbd_put_io(struct thread_data *td, const struct io_u *io_u)
 {
 	const struct fio_file *f = io_u->file;
 	struct zoned_block_device_info *zbd_info = f->zbd_info;
 	struct fio_zone_info *z;
 	uint32_t zone_idx;
-	int ret;
 
-	if (!zbd_info)
-		return;
+	assert(zbd_info);
 
 	zone_idx = zbd_zone_idx(f, io_u->offset);
 	assert(zone_idx < zbd_info->nr_zones);
-	z = &zbd_info->zone_info[zone_idx];
+	z = get_zone(f, zone_idx);
 
-	if (!zbd_zone_swr(z))
-		return;
+	assert(z->has_wp);
 
 	dprint(FD_ZBD,
 	       "%s: terminate I/O (%lld, %llu) for zone %u\n",
 	       f->file_name, io_u->offset, io_u->buflen, zone_idx);
 
-	ret = pthread_mutex_unlock(&z->mutex);
-	assert(ret == 0);
-	zbd_check_swd(f);
+	zbd_end_zone_io(td, io_u, z);
+
+	zone_unlock(z);
+	zbd_check_swd(td, f);
 }
 
 /*
@@ -1293,6 +1585,28 @@ void setup_zbd_zone_mode(struct thread_data *td, struct io_u *io_u)
 
 	assert(td->o.zone_mode == ZONE_MODE_ZBD);
 	assert(td->o.zone_size);
+	assert(f->zbd_info);
+
+	zone_idx = zbd_zone_idx(f, f->last_pos[ddir]);
+	z = get_zone(f, zone_idx);
+
+	/*
+	 * When the zone capacity is smaller than the zone size and the I/O is
+	 * sequential write, skip to zone end if the latest position is at the
+	 * zone capacity limit.
+	 */
+	if (z->capacity < f->zbd_info->zone_size && !td_random(td) &&
+	    ddir == DDIR_WRITE &&
+	    f->last_pos[ddir] >= zbd_zone_capacity_end(z)) {
+		dprint(FD_ZBD,
+		       "%s: Jump from zone capacity limit to zone end:"
+		       " (%llu -> %llu) for zone %u (%llu)\n",
+		       f->file_name, (unsigned long long) f->last_pos[ddir],
+		       (unsigned long long) zbd_zone_end(z), zone_idx,
+		       (unsigned long long) z->capacity);
+		td->io_skip_bytes += zbd_zone_end(z) - f->last_pos[ddir];
+		f->last_pos[ddir] = zbd_zone_end(z);
+	}
 
 	/*
 	 * zone_skip is valid only for sequential workloads.
@@ -1307,11 +1621,8 @@ void setup_zbd_zone_mode(struct thread_data *td, struct io_u *io_u)
 	 * - For reads with td->o.read_beyond_wp == false, the last position
 	 *   reached the zone write pointer.
 	 */
-	zone_idx = zbd_zone_idx(f, f->last_pos[ddir]);
-	z = &f->zbd_info->zone_info[zone_idx];
-
 	if (td->zone_bytes >= td->o.zone_size ||
-	    f->last_pos[ddir] >= (z+1)->start ||
+	    f->last_pos[ddir] >= zbd_zone_end(z) ||
 	    (ddir == DDIR_READ &&
 	     (!td->o.read_beyond_wp) && f->last_pos[ddir] >= z->wp)) {
 		/*
@@ -1332,7 +1643,7 @@ void setup_zbd_zone_mode(struct thread_data *td, struct io_u *io_u)
 }
 
 /**
- * zbd_adjust_ddir - Adjust an I/O direction for zonedmode=zbd.
+ * zbd_adjust_ddir - Adjust an I/O direction for zonemode=zbd.
  *
  * @td: FIO thread data.
  * @io_u: FIO I/O unit.
@@ -1349,9 +1660,8 @@ enum fio_ddir zbd_adjust_ddir(struct thread_data *td, struct io_u *io_u,
 	 * devices with all empty zones. Overwrite the first I/O direction as
 	 * write to make sure data to read exists.
 	 */
-	if (td->o.zone_mode != ZONE_MODE_ZBD ||
-	    ddir != DDIR_READ ||
-	    !td_rw(td))
+	assert(io_u->file->zbd_info);
+	if (ddir != DDIR_READ || !td_rw(td))
 		return ddir;
 
 	if (io_u->file->zbd_info->sectors_with_data ||
@@ -1380,19 +1690,39 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
 	uint64_t new_len;
 	int64_t range;
 
-	if (!f->zbd_info)
-		return io_u_accept;
-
+	assert(f->zbd_info);
 	assert(min_bs);
 	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];
+	zb = get_zone(f, zone_idx_b);
 	orig_zb = zb;
 
-	/* Accept the I/O offset for conventional zones. */
-	if (!zbd_zone_swr(zb))
+	if (!zb->has_wp) {
+		/* Accept non-write I/Os for conventional zones. */
+		if (io_u->ddir != DDIR_WRITE)
+			return io_u_accept;
+		/*
+		 * Make sure that writes to conventional zones
+		 * don't cross over to any sequential zones.
+		 */
+		if (!(zb + 1)->has_wp ||
+		    io_u->offset + io_u->buflen <= (zb + 1)->start)
+			return io_u_accept;
+
+		if (io_u->offset + min_bs > (zb + 1)->start) {
+			dprint(FD_IO,
+			       "%s: off=%llu + min_bs=%u > next zone %llu\n",
+			       f->file_name, io_u->offset,
+			       min_bs, (unsigned long long) (zb + 1)->start);
+			io_u->offset = zb->start + (zb + 1)->start - io_u->offset;
+			new_len = min(io_u->buflen, (zb + 1)->start - io_u->offset);
+		} else {
+			new_len = (zb + 1)->start - io_u->offset;
+		}
+		io_u->buflen = new_len / min_bs * min_bs;
 		return io_u_accept;
+	}
 
 	/*
 	 * Accept the I/O offset for reads if reading beyond the write pointer
@@ -1402,13 +1732,13 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
 	    io_u->ddir == DDIR_READ && td->o.read_beyond_wp)
 		return io_u_accept;
 
-	zbd_check_swd(f);
+	zbd_check_swd(td, f);
 
-	zone_lock(td, zb);
+	zone_lock(td, f, zb);
 
 	switch (io_u->ddir) {
 	case DDIR_READ:
-		if (td->runstate == TD_VERIFYING) {
+		if (td->runstate == TD_VERIFYING && td_write(td)) {
 			zb = zbd_replay_write_order(td, io_u, zb);
 			goto accept;
 		}
@@ -1421,9 +1751,8 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
 			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)];
+			zone_unlock(zb);
+			zl = get_zone(f, f->max_zone);
 			zb = zbd_find_zone(td, io_u, zb, zl);
 			if (!zb) {
 				dprint(FD_ZBD,
@@ -1452,6 +1781,12 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
 			io_u->offset = zb->start +
 				((io_u->offset - orig_zb->start) %
 				 (range - io_u->buflen)) / min_bs * min_bs;
+		/*
+		 * When zbd_find_zone() returns a conventional zone,
+		 * we can simply accept the new i/o offset here.
+		 */
+		if (!zb->has_wp)
+			return io_u_accept;
 		/*
 		 * Make sure the I/O does not cross over the zone wp position.
 		 */
@@ -1467,18 +1802,27 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
 		assert(io_u->offset + io_u->buflen <= zb->wp);
 		goto accept;
 	case DDIR_WRITE:
-		if (io_u->buflen > f->zbd_info->zone_size)
+		if (io_u->buflen > f->zbd_info->zone_size) {
+			td_verror(td, EINVAL, "I/O buflen exceeds zone size");
+			dprint(FD_IO,
+			       "%s: I/O buflen %llu exceeds zone size %llu\n",
+			       f->file_name, io_u->buflen,
+			       (unsigned long long) f->zbd_info->zone_size);
 			goto eof;
-		if (!zbd_open_zone(td, io_u, zone_idx_b)) {
-			pthread_mutex_unlock(&zb->mutex);
+		}
+		if (!zbd_open_zone(td, f, zone_idx_b)) {
+			zone_unlock(zb);
 			zb = zbd_convert_to_open_zone(td, io_u);
-			if (!zb)
+			if (!zb) {
+				dprint(FD_IO, "%s: can't convert to open zone",
+				       f->file_name);
 				goto eof;
-			zone_idx_b = zb - f->zbd_info->zone_info;
+			}
+			zone_idx_b = zbd_zone_nr(f, zb);
 		}
 		/* Check whether the zone reset threshold has been exceeded */
 		if (td->o.zrf.u.f) {
-			if (f->zbd_info->sectors_with_data >=
+			if (f->zbd_info->wp_sectors_with_data >=
 			    f->io_size * td->o.zrt.u.f &&
 			    zbd_dec_and_reset_write_cnt(td, f)) {
 				zb->reset_zone = 1;
@@ -1498,13 +1842,22 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
 			zb->reset_zone = 0;
 			if (zbd_reset_zone(td, f, zb) < 0)
 				goto eof;
+
+			if (zb->capacity < min_bs) {
+				td_verror(td, EINVAL, "ZCAP is less min_bs");
+				log_err("zone capacity %llu smaller than minimum block size %d\n",
+					(unsigned long long)zb->capacity,
+					min_bs);
+				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);
+			td_verror(td, EINVAL, "invalid WP value");
+			dprint(FD_ZBD, "%s: dropped request with offset %llu\n",
+			       f->file_name, io_u->offset);
 			goto eof;
 		}
 		/*
@@ -1513,7 +1866,7 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
 		 * small.
 		 */
 		new_len = min((unsigned long long)io_u->buflen,
-			      (zb + 1)->start - io_u->offset);
+			      zbd_zone_capacity_end(zb) - io_u->offset);
 		new_len = new_len / min_bs * min_bs;
 		if (new_len == io_u->buflen)
 			goto accept;
@@ -1523,9 +1876,9 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
 			       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);
+		td_verror(td, EIO, "zone remainder too small");
+		log_err("zone remainder %lld smaller than min block size %d\n",
+			(zbd_zone_capacity_end(zb) - io_u->offset), min_bs);
 		goto eof;
 	case DDIR_TRIM:
 		/* fall-through */
@@ -1541,17 +1894,23 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
 	assert(false);
 
 accept:
-	assert(zb);
+	assert(zb->has_wp);
 	assert(zb->cond != ZBD_ZONE_COND_OFFLINE);
 	assert(!io_u->zbd_queue_io);
 	assert(!io_u->zbd_put_io);
 	io_u->zbd_queue_io = zbd_queue_io;
 	io_u->zbd_put_io = zbd_put_io;
+	/*
+	 * Since we return with the zone lock still held,
+	 * add an annotation to let Coverity know that it
+	 * is intentional.
+	 */
+	/* coverity[missing_unlock] */
 	return io_u_accept;
 
 eof:
-	if (zb)
-		pthread_mutex_unlock(&zb->mutex);
+	if (zb && zb->has_wp)
+		zone_unlock(zb);
 	return io_u_eof;
 }