X-Git-Url: https://git.kernel.dk/?a=blobdiff_plain;f=zbd.c;h=f2599bd4871cd97e7028654d04083fd5eba64d90;hb=59f94d26f98e9c0bc18d4e013f3361c51a2c6b25;hp=a64fd0c710bcf98b0df121b548e01f5ef4a4b3f6;hpb=767d1372a2f045eba88f0520bdbe0a1e250f7f79;p=fio.git

diff --git a/zbd.c b/zbd.c
index a64fd0c7..f2599bd4 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"
 
 /**
@@ -139,6 +140,24 @@ static inline bool zbd_zone_swr(struct fio_zone_info *z)
 	return z->type == ZBD_ZONE_TYPE_SWR;
 }
 
+/**
+ * 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 uint64_t zbd_zone_capacity_end(const struct fio_zone_info *z)
+{
+	return z->start + z->capacity;
+}
+
 /**
  * zbd_zone_full - verify whether a minimum number of bytes remain in a zone
  * @f: file pointer.
@@ -153,11 +172,17 @@ 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;
+		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, 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);
+
 	/*
 	 * 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.
@@ -262,8 +287,9 @@ 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;
+			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);
@@ -278,7 +304,8 @@ static bool zbd_verify_sizes(void)
 			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 ((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);
@@ -289,6 +316,10 @@ static bool zbd_verify_sizes(void)
 					 (unsigned long long) new_end - f->file_offset);
 				f->io_size = new_end - f->file_offset;
 			}
+
+			f->min_zone = zbd_zone_idx(f, f->file_offset);
+			f->max_zone = zbd_zone_idx(f, f->file_offset + f->io_size);
+			assert(f->min_zone < f->max_zone);
 		}
 	}
 
@@ -307,7 +338,7 @@ static bool zbd_verify_bs(void)
 			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",
@@ -341,8 +372,8 @@ 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) {
@@ -357,24 +388,33 @@ 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;
+	}
+
 	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;
 	}
 	/* a sentinel */
 	p->start = nr_zones * zone_size;
@@ -384,7 +424,6 @@ 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;
 }
 
@@ -404,13 +443,8 @@ 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);
-
 	zones = calloc(ZBD_REPORT_MAX_ZONES, sizeof(struct zbd_zone));
 	if (!zones)
 		goto out;
@@ -444,18 +478,20 @@ static int parse_zone_info(struct thread_data *td, struct fio_file *f)
 	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);
@@ -501,7 +537,6 @@ 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;
 }
 
@@ -537,8 +572,10 @@ static int zbd_create_zone_info(struct thread_data *td, struct fio_file *f)
 		return -EINVAL;
 	}
 
-	if (ret == 0)
+	if (ret == 0) {
 		f->zbd_info->model = zbd_model;
+		f->zbd_info->max_open_zones = td->o.max_open_zones;
+	}
 	return ret;
 }
 
@@ -546,8 +583,7 @@ 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 +628,12 @@ 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_setup_files(struct thread_data *td)
 {
 	struct fio_file *f;
 	int i;
@@ -613,55 +654,45 @@ int zbd_init(struct thread_data *td)
 	if (!zbd_verify_bs())
 		return 1;
 
-	return 0;
-}
+	for_each_file(td, f, i) {
+		struct zoned_block_device_info *zbd = f->zbd_info;
+		struct fio_zone_info *z;
+		int zi;
 
-/**
- * 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, struct fio_file *f,
-			   uint64_t offset, uint64_t length)
-{
-	uint32_t zone_idx_b, zone_idx_e;
-	struct fio_zone_info *zb, *ze, *z;
-	int ret = 0;
+		if (!zbd)
+			continue;
 
-	assert(is_valid_offset(f, offset + length - 1));
+		zbd->max_open_zones = zbd->max_open_zones ?: ZBD_MAX_OPEN_ZONES;
 
-	switch (f->zbd_info->model) {
-	case ZBD_HOST_AWARE:
-	case ZBD_HOST_MANAGED:
-		ret = zbd_reset_wp(td, f, offset, length);
-		if (ret < 0)
-			return ret;
-		break;
-	default:
-		break;
-	}
+		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;
+		}
 
-	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);
+		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;
+			}
+		}
 	}
 
-	td->ts.nr_zone_resets += ze - zb;
-
-	return ret;
+	return 0;
 }
 
 static unsigned int zbd_zone_nr(struct zoned_block_device_info *zbd_info,
@@ -677,29 +708,68 @@ static unsigned int zbd_zone_nr(struct zoned_block_device_info *zbd_info,
  * @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_zone(struct thread_data *td, struct fio_file *f,
 			  struct fio_zone_info *z)
 {
+	uint64_t offset = z->start;
+	uint64_t length = (z+1)->start - offset;
+	int ret = 0;
+
+	if (z->wp == z->start)
+		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->zbd_info, z));
+	switch (f->zbd_info->model) {
+	case ZBD_HOST_AWARE:
+	case ZBD_HOST_MANAGED:
+		ret = zbd_reset_wp(td, f, offset, length);
+		if (ret < 0)
+			return ret;
+		break;
+	default:
+		break;
+	}
 
-	return zbd_reset_range(td, f, z->start, (z+1)->start - z->start);
+	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;
+
+	td->ts.nr_zone_resets++;
+
+	return ret;
 }
 
 /* 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) {
+		dprint(FD_ZBD, "%s: zone %d is not open\n",
+		       f->file_name, zone_idx);
+		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--;
+	td->num_open_zones--;
 	f->zbd_info->zone_info[zone_idx].open = 0;
 }
 
@@ -730,15 +800,10 @@ static int zbd_reset_zones(struct thread_data *td, struct fio_file *f,
 
 		if (!zbd_zone_swr(z))
 			continue;
-		zone_lock(td, z);
+		zone_lock(td, f, 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);
-			}
+			zbd_close_zone(td, f, nz);
 			pthread_mutex_unlock(&f->zbd_info->mutex);
 
 			reset_wp = z->wp != z->start;
@@ -762,14 +827,20 @@ 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);
 }
 
@@ -783,7 +854,7 @@ 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;
@@ -800,9 +871,8 @@ 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)];
+	zb = &f->zbd_info->zone_info[f->min_zone];
+	ze = &f->zbd_info->zone_info[f->max_zone];
 	for (z = zb; z < ze; z++) {
 		pthread_mutex_lock(&z->mutex);
 		swd += z->wp - z->start;
@@ -853,14 +923,12 @@ 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;
-	uint32_t zone_idx_e;
 
 	if (!f->zbd_info || !td_write(td))
 		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];
+	zb = &f->zbd_info->zone_info[f->min_zone];
+	ze = &f->zbd_info->zone_info[f->max_zone];
 	zbd_init_swd(f);
 	/*
 	 * If data verification is enabled reset the affected zones before
@@ -879,8 +947,9 @@ 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)
@@ -895,11 +964,10 @@ 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];
 	bool res = true;
 
@@ -913,18 +981,26 @@ 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;
 
@@ -951,15 +1027,16 @@ 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.
@@ -968,6 +1045,10 @@ 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);
 
@@ -982,16 +1063,14 @@ static struct fio_zone_info *zbd_convert_to_open_zone(struct thread_data *td,
 
 		z = &f->zbd_info->zone_info[zone_idx];
 
-		zone_lock(td, z);
+		zone_lock(td, f, z);
 		pthread_mutex_lock(&f->zbd_info->mutex);
-		if (td->o.max_open_zones == 0)
+		if (td->o.max_open_zones == 0 && td->o.job_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;
+			goto open_other_zone;
 		}
 
 		/*
@@ -1008,8 +1087,7 @@ 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;
 			}
@@ -1035,18 +1113,34 @@ found_candidate_zone:
 	/* 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++;
@@ -1054,14 +1148,14 @@ examine_zone:
 		z++;
 		if (!is_valid_offset(f, z->start)) {
 			/* Wrap-around. */
-			zone_idx = zbd_zone_idx(f, f->file_offset);
+			zone_idx = f->min_zone;
 			z = &f->zbd_info->zone_info[zone_idx];
 		}
 		assert(is_valid_offset(f, z->start));
-		zone_lock(td, z);
+		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;
 	}
 
@@ -1071,13 +1165,15 @@ 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);
 
 		z = &f->zbd_info->zone_info[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);
 	}
@@ -1102,15 +1198,15 @@ 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)) {
+	if (!zbd_open_zone(td, f, 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)
+	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);
+			min_bs, (unsigned long long)z->capacity);
 	io_u->offset = z->start + z->verify_block++ * min_bs;
 	return z;
 }
@@ -1128,10 +1224,10 @@ 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)];
+		&f->zbd_info->zone_info[f->min_zone];
 
 	/*
 	 * Skip to the next non-empty zone in case of sequential I/O and to
@@ -1139,7 +1235,7 @@ 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) {
-			zone_lock(td, z1);
+			zone_lock(td, f, z1);
 			if (z1->start + min_bs <= z1->wp)
 				return z1;
 			pthread_mutex_unlock(&z1->mutex);
@@ -1148,7 +1244,7 @@ zbd_find_zone(struct thread_data *td, struct io_u *io_u,
 		}
 		if (td_random(td) && z2 >= zf &&
 		    z2->cond != ZBD_ZONE_COND_OFFLINE) {
-			zone_lock(td, z2);
+			zone_lock(td, f, z2);
 			if (z2->start + min_bs <= z2->wp)
 				return z2;
 			pthread_mutex_unlock(&z2->mutex);
@@ -1159,6 +1255,28 @@ 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, z - f->zbd_info->zone_info);
+		pthread_mutex_unlock(&f->zbd_info->mutex);
+	}
+}
+
 /**
  * zbd_queue_io - update the write pointer of a sequential zone
  * @io_u: I/O unit
@@ -1168,7 +1286,8 @@ 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;
@@ -1196,7 +1315,7 @@ 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
@@ -1214,6 +1333,9 @@ 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 */
@@ -1226,7 +1348,7 @@ 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;
@@ -1248,6 +1370,8 @@ static void zbd_put_io(const struct io_u *io_u)
 	       "%s: terminate I/O (%lld, %llu) for zone %u\n",
 	       f->file_name, io_u->offset, io_u->buflen, zone_idx);
 
+	zbd_end_zone_io(td, io_u, z);
+
 	ret = pthread_mutex_unlock(&z->mutex);
 	assert(ret == 0);
 	zbd_check_swd(f);
@@ -1292,6 +1416,28 @@ 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);
 
+	zone_idx = zbd_zone_idx(f, f->last_pos[ddir]);
+	z = &f->zbd_info->zone_info[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),
+		       zbd_zone_nr(f->zbd_info, z),
+		       (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.
 	 */
@@ -1305,11 +1451,8 @@ 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)) {
 		/*
@@ -1330,7 +1473,7 @@ 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.
@@ -1347,9 +1490,7 @@ 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))
+	if (ddir != DDIR_READ || !td_rw(td))
 		return ddir;
 
 	if (io_u->file->zbd_info->sectors_with_data ||
@@ -1402,13 +1543,13 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
 
 	zbd_check_swd(f);
 
-	zone_lock(td, zb);
+	zone_lock(td, f, zb);
 
 	switch (io_u->ddir) {
 	case DDIR_READ:
-		if (td->runstate == TD_VERIFYING) {
-			if (td_write(td))
-				zb = zbd_replay_write_order(td, io_u, zb);
+		if (td->runstate == TD_VERIFYING && td_write(td)) {
+			zb = zbd_replay_write_order(td, io_u, zb);
+			pthread_mutex_unlock(&zb->mutex);
 			goto accept;
 		}
 		/*
@@ -1421,8 +1562,7 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
 		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)];
+			zl = &f->zbd_info->zone_info[f->max_zone];
 			zb = zbd_find_zone(td, io_u, zb, zl);
 			if (!zb) {
 				dprint(FD_ZBD,
@@ -1468,7 +1608,7 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
 	case DDIR_WRITE:
 		if (io_u->buflen > f->zbd_info->zone_size)
 			goto eof;
-		if (!zbd_open_zone(td, io_u, zone_idx_b)) {
+		if (!zbd_open_zone(td, f, zone_idx_b)) {
 			pthread_mutex_unlock(&zb->mutex);
 			zb = zbd_convert_to_open_zone(td, io_u);
 			if (!zb)
@@ -1497,6 +1637,13 @@ 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) {
+				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));
@@ -1512,7 +1659,7 @@ 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;
@@ -1523,7 +1670,7 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
 			goto accept;
 		}
 		log_err("Zone remainder %lld smaller than minimum block size %d\n",
-			((zb + 1)->start - io_u->offset),
+			(zbd_zone_capacity_end(zb) - io_u->offset),
 			min_bs);
 		goto eof;
 	case DDIR_TRIM: