return (uint64_t)(offset - f->file_offset) < f->io_size;
}
-static inline unsigned int zbd_zone_nr(const struct fio_file *f,
- struct fio_zone_info *zone)
+static inline unsigned int zbd_zone_idx(const struct fio_file *f,
+ struct fio_zone_info *zone)
{
return zone - f->zbd_info->zone_info;
}
/**
- * zbd_zone_idx - convert an offset into a zone number
+ * zbd_offset_to_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)
+static unsigned int zbd_offset_to_zone_idx(const struct fio_file *f,
+ uint64_t offset)
{
uint32_t zone_idx;
return z->start + z->capacity;
}
+/**
+ * zbd_zone_remainder - Return the number of bytes that are still available for
+ * writing before the zone gets full
+ * @z: zone info pointer.
+ */
+static inline uint64_t zbd_zone_remainder(struct fio_zone_info *z)
+{
+ if (z->wp >= zbd_zone_capacity_end(z))
+ return 0;
+
+ return zbd_zone_capacity_end(z) - z->wp;
+}
+
/**
* zbd_zone_full - verify whether a minimum number of bytes remain in a zone
* @f: file pointer.
{
assert((required & 511) == 0);
- return z->has_wp &&
- z->wp + required > zbd_zone_capacity_end(z);
+ return z->has_wp && required > zbd_zone_remainder(z);
}
static void zone_lock(struct thread_data *td, const struct fio_file *f,
assert(!ret);
}
-static inline struct fio_zone_info *get_zone(const struct fio_file *f,
- unsigned int zone_nr)
+static inline struct fio_zone_info *zbd_get_zone(const struct fio_file *f,
+ unsigned int zone_idx)
+{
+ return &f->zbd_info->zone_info[zone_idx];
+}
+
+static inline struct fio_zone_info *
+zbd_offset_to_zone(const struct fio_file *f, uint64_t offset)
{
- return &f->zbd_info->zone_info[zone_nr];
+ return zbd_get_zone(f, zbd_offset_to_zone_idx(f, offset));
}
/**
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));
+ f->file_name, zbd_zone_idx(f, z));
switch (f->zbd_info->model) {
case ZBD_HOST_AWARE:
pthread_mutex_unlock(&f->zbd_info->mutex);
z->wp = z->start;
- z->verify_block = 0;
td->ts.nr_zone_resets++;
return;
for (ozi = 0; ozi < f->zbd_info->num_open_zones; ozi++) {
- if (get_zone(f, f->zbd_info->open_zones[ozi]) == z)
+ if (zbd_get_zone(f, f->zbd_info->open_zones[ozi]) == z)
break;
}
if (ozi == f->zbd_info->num_open_zones)
return;
dprint(FD_ZBD, "%s: closing zone %u\n",
- f->file_name, zbd_zone_nr(f, z));
+ f->file_name, zbd_zone_idx(f, z));
memmove(f->zbd_info->open_zones + ozi,
f->zbd_info->open_zones + ozi + 1,
z->open = 0;
}
+/**
+ * zbd_finish_zone - finish the specified zone
+ * @td: FIO thread data.
+ * @f: FIO file for which to finish a zone
+ * @z: Zone to finish.
+ *
+ * Finish the zone at @offset with open or close status.
+ */
+static int zbd_finish_zone(struct thread_data *td, struct fio_file *f,
+ struct fio_zone_info *z)
+{
+ uint64_t offset = z->start;
+ uint64_t length = f->zbd_info->zone_size;
+ int ret = 0;
+
+ switch (f->zbd_info->model) {
+ case ZBD_HOST_AWARE:
+ case ZBD_HOST_MANAGED:
+ if (td->io_ops && td->io_ops->finish_zone)
+ ret = td->io_ops->finish_zone(td, f, offset, length);
+ else
+ ret = blkzoned_finish_zone(td, f, offset, length);
+ break;
+ default:
+ break;
+ }
+
+ if (ret < 0) {
+ td_verror(td, errno, "finish zone failed");
+ log_err("%s: finish zone at sector %"PRIu64" failed (%d).\n",
+ f->file_name, offset >> 9, errno);
+ } else {
+ z->wp = (z+1)->start;
+ }
+
+ return ret;
+}
+
/**
* zbd_reset_zones - Reset a range of zones.
* @td: fio thread data.
assert(min_bs);
dprint(FD_ZBD, "%s: examining zones %u .. %u\n",
- f->file_name, zbd_zone_nr(f, zb), zbd_zone_nr(f, ze));
+ f->file_name, zbd_zone_idx(f, zb), zbd_zone_idx(f, ze));
for (z = zb; z < ze; z++) {
if (!z->has_wp)
if (z->wp != z->start) {
dprint(FD_ZBD, "%s: resetting zone %u\n",
- f->file_name, zbd_zone_nr(f, z));
+ f->file_name, zbd_zone_idx(f, z));
if (zbd_reset_zone(td, f, z) < 0)
res = 1;
}
{
const uint64_t min_bs = td->o.min_bs[DDIR_WRITE];
struct zoned_block_device_info *zbdi = f->zbd_info;
- uint32_t zone_idx = zbd_zone_nr(f, z);
+ uint32_t zone_idx = zbd_zone_idx(f, z);
bool res = true;
if (z->cond == ZBD_ZONE_COND_OFFLINE)
* already in-flight, handle it as a full zone instead of an
* open zone.
*/
- if (z->wp >= zbd_zone_capacity_end(z))
+ if (!zbd_zone_remainder(z))
res = false;
goto out;
}
return res;
}
-/* Verify whether direct I/O is used for all host-managed zoned drives. */
+/* Verify whether direct I/O is used for all host-managed zoned block drives. */
static bool zbd_using_direct_io(void)
{
struct thread_data *td;
if (td->o.odirect || !(td->o.td_ddir & TD_DDIR_WRITE))
continue;
for_each_file(td, f, j) {
- if (f->zbd_info &&
+ if (f->zbd_info && f->filetype == FIO_TYPE_BLOCK &&
f->zbd_info->model == ZBD_HOST_MANAGED)
return false;
}
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);
+ zone_idx_b = zbd_offset_to_zone_idx(f, f->file_offset);
+ zone_idx_e =
+ zbd_offset_to_zone_idx(f, f->file_offset + f->io_size - 1);
for (zone_idx = zone_idx_b; zone_idx <= zone_idx_e; zone_idx++)
- if (get_zone(f, zone_idx)->has_wp)
+ if (zbd_get_zone(f, zone_idx)->has_wp)
return true;
return false;
{
const struct fio_zone_info *z;
uint64_t new_offset, new_end;
- uint32_t zone_idx;
if (!f->zbd_info)
return true;
return false;
}
- zone_idx = zbd_zone_idx(f, f->file_offset);
- z = get_zone(f, zone_idx);
+ z = zbd_offset_to_zone(f, f->file_offset);
if ((f->file_offset != z->start) &&
(td->o.td_ddir != TD_DDIR_READ)) {
new_offset = zbd_zone_end(z);
f->file_offset = new_offset;
}
- zone_idx = zbd_zone_idx(f, f->file_offset + f->io_size);
- z = get_zone(f, zone_idx);
+ z = zbd_offset_to_zone(f, f->file_offset + f->io_size);
new_end = z->start;
if ((td->o.td_ddir != TD_DDIR_READ) &&
(f->file_offset + f->io_size != new_end)) {
{
struct thread_data *td;
struct fio_file *f;
- int i, j, k;
+ int i, j;
for_each_td(td, i) {
if (td_trim(td) &&
zone_size);
return false;
}
- 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 %"PRIu64"\n",
- f->file_name, td->o.bs[k],
- zone_size);
- return false;
- }
- }
}
}
return true;
if (!zbd_verify_bs())
return 1;
+ if (td->o.experimental_verify) {
+ log_err("zonemode=zbd does not support experimental verify\n");
+ return 1;
+ }
+
for_each_file(td, f, i) {
struct zoned_block_device_info *zbd = f->zbd_info;
struct fio_zone_info *z;
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);
+ f->min_zone = zbd_offset_to_zone_idx(f, f->file_offset);
+ f->max_zone =
+ zbd_offset_to_zone_idx(f, f->file_offset + f->io_size);
/*
* When all zones in the I/O range are conventional, io_size
uint64_t swd = 0;
uint64_t wp_swd = 0;
- zb = get_zone(f, f->min_zone);
- ze = get_zone(f, f->max_zone);
+ zb = zbd_get_zone(f, f->min_zone);
+ ze = zbd_get_zone(f, f->max_zone);
for (z = zb; z < ze; z++) {
if (z->has_wp) {
zone_lock(td, f, z);
{
struct fio_zone_info *zb, *ze;
uint64_t swd;
+ bool verify_data_left = false;
if (!f->zbd_info || !td_write(td))
return;
- zb = get_zone(f, f->min_zone);
- ze = get_zone(f, f->max_zone);
+ zb = zbd_get_zone(f, f->min_zone);
+ ze = zbd_get_zone(f, f->max_zone);
swd = zbd_process_swd(td, f, SET_SWD);
dprint(FD_ZBD, "%s(%s): swd = %" PRIu64 "\n",
* writing any data to avoid that a zone reset has to be issued while
* writing data, which causes data loss.
*/
- if (td->o.verify != VERIFY_NONE && td->runstate != TD_VERIFYING)
- zbd_reset_zones(td, f, zb, ze);
+ if (td->o.verify != VERIFY_NONE) {
+ verify_data_left = td->runstate == TD_VERIFYING ||
+ td->io_hist_len || td->verify_batch;
+ if (td->io_hist_len && td->o.verify_backlog)
+ verify_data_left =
+ td->io_hist_len % td->o.verify_backlog;
+ if (!verify_data_left)
+ zbd_reset_zones(td, f, zb, ze);
+ }
+
zbd_reset_write_cnt(td, f);
}
*/
zone_idx = zbdi->open_zones[pick_random_zone_idx(f, io_u)];
} else {
- zone_idx = zbd_zone_idx(f, io_u->offset);
+ zone_idx = zbd_offset_to_zone_idx(f, io_u->offset);
}
if (zone_idx < f->min_zone)
zone_idx = f->min_zone;
for (;;) {
uint32_t tmp_idx;
- z = get_zone(f, zone_idx);
+ z = zbd_get_zone(f, zone_idx);
if (z->has_wp)
zone_lock(td, f, z);
/* Both z->mutex and zbdi->mutex are held. */
examine_zone:
- if (z->wp + min_bs <= zbd_zone_capacity_end(z)) {
+ if (zbd_zone_remainder(z) >= min_bs) {
pthread_mutex_unlock(&zbdi->mutex);
goto out;
}
if (!is_valid_offset(f, z->start)) {
/* Wrap-around. */
zone_idx = f->min_zone;
- z = get_zone(f, zone_idx);
+ z = zbd_get_zone(f, zone_idx);
}
assert(is_valid_offset(f, z->start));
if (!z->has_wp)
pthread_mutex_unlock(&zbdi->mutex);
zone_unlock(z);
- z = get_zone(f, zone_idx);
+ z = zbd_get_zone(f, zone_idx);
zone_lock(td, f, z);
- if (z->wp + min_bs <= zbd_zone_capacity_end(z))
+ if (zbd_zone_remainder(z) >= min_bs)
goto out;
pthread_mutex_lock(&zbdi->mutex);
}
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 uint64_t min_bs = td->o.min_bs[DDIR_WRITE];
-
- if (!zbd_open_zone(td, f, z)) {
- zone_unlock(z);
- z = zbd_convert_to_open_zone(td, io_u);
- assert(z);
- }
-
- if (z->verify_block * min_bs >= z->capacity) {
- log_err("%s: %d * %"PRIu64" >= %"PRIu64"\n",
- f->file_name, z->verify_block, min_bs, 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 >= %"PRIu64"\n",
- f->file_name, io_u->offset, io_u->buflen,
- zbd_zone_capacity_end(z));
- assert(false);
- }
- z->verify_block += io_u->buflen / min_bs;
-
- return z;
-}
-
/*
* Find another zone which has @min_bytes of readable data. Search in zones
* @zb + 1 .. @zl. For random workload, also search in zones @zb - 1 .. @zf.
{
struct fio_file *f = io_u->file;
struct fio_zone_info *z1, *z2;
- const struct fio_zone_info *const zf = get_zone(f, f->min_zone);
+ const struct fio_zone_info *const zf = zbd_get_zone(f, f->min_zone);
/*
* Skip to the next non-empty zone in case of sequential I/O and to
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;
uint64_t zone_end;
assert(zbd_info);
- zone_idx = zbd_zone_idx(f, io_u->offset);
- assert(zone_idx < zbd_info->nr_zones);
- z = get_zone(f, zone_idx);
-
+ z = zbd_offset_to_zone(f, io_u->offset);
assert(z->has_wp);
if (!success)
dprint(FD_ZBD,
"%s: queued I/O (%lld, %llu) for zone %u\n",
- f->file_name, io_u->offset, io_u->buflen, zone_idx);
+ f->file_name, io_u->offset, io_u->buflen, zbd_zone_idx(f, z));
switch (io_u->ddir) {
case DDIR_WRITE:
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;
assert(zbd_info);
- zone_idx = zbd_zone_idx(f, io_u->offset);
- assert(zone_idx < zbd_info->nr_zones);
- z = get_zone(f, zone_idx);
-
+ z = zbd_offset_to_zone(f, io_u->offset);
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);
+ f->file_name, io_u->offset, io_u->buflen, zbd_zone_idx(f, z));
zbd_end_zone_io(td, io_u, z);
struct fio_file *f = io_u->file;
enum fio_ddir ddir = io_u->ddir;
struct fio_zone_info *z;
- uint32_t zone_idx;
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);
+ z = zbd_offset_to_zone(f, f->last_pos[ddir]);
/*
* When the zone capacity is smaller than the zone size and the I/O is
"%s: Jump from zone capacity limit to zone end:"
" (%"PRIu64" -> %"PRIu64") for zone %u (%"PRIu64")\n",
f->file_name, f->last_pos[ddir],
- zbd_zone_end(z), zone_idx, z->capacity);
+ zbd_zone_end(z), zbd_zone_idx(f, z), z->capacity);
td->io_skip_bytes += zbd_zone_end(z) - f->last_pos[ddir];
f->last_pos[ddir] = zbd_zone_end(z);
}
{
struct fio_file *f = io_u->file;
struct zoned_block_device_info *zbdi = f->zbd_info;
- uint32_t zone_idx_b;
struct fio_zone_info *zb, *zl, *orig_zb;
uint32_t orig_len = io_u->buflen;
uint64_t min_bs = td->o.min_bs[io_u->ddir];
assert(is_valid_offset(f, io_u->offset));
assert(io_u->buflen);
- zone_idx_b = zbd_zone_idx(f, io_u->offset);
- zb = get_zone(f, zone_idx_b);
+ zb = zbd_offset_to_zone(f, io_u->offset);
orig_zb = zb;
if (!zb->has_wp) {
switch (io_u->ddir) {
case DDIR_READ:
- if (td->runstate == TD_VERIFYING && td_write(td)) {
- zb = zbd_replay_write_order(td, io_u, zb);
+ if (td->runstate == TD_VERIFYING && td_write(td))
goto accept;
- }
/*
* Check that there is enough written data in the zone to do an
if (range < min_bs ||
((!td_random(td)) && (io_u->offset + min_bs > zb->wp))) {
zone_unlock(zb);
- zl = get_zone(f, f->max_zone);
+ zl = zbd_get_zone(f, f->max_zone);
zb = zbd_find_zone(td, io_u, min_bs, zb, zl);
if (!zb) {
dprint(FD_ZBD,
goto eof;
}
+retry:
+ if (zbd_zone_remainder(zb) > 0 &&
+ zbd_zone_remainder(zb) < min_bs) {
+ pthread_mutex_lock(&f->zbd_info->mutex);
+ zbd_close_zone(td, f, zb);
+ pthread_mutex_unlock(&f->zbd_info->mutex);
+ dprint(FD_ZBD,
+ "%s: finish zone %d\n",
+ f->file_name, zbd_zone_idx(f, zb));
+ io_u_quiesce(td);
+ zbd_finish_zone(td, f, zb);
+ if (zbd_zone_idx(f, zb) + 1 >= f->max_zone) {
+ if (!td_random(td))
+ goto eof;
+ }
+ zone_unlock(zb);
+
+ /* Find the next write pointer zone */
+ do {
+ zb++;
+ if (zbd_zone_idx(f, zb) >= f->max_zone)
+ zb = zbd_get_zone(f, f->min_zone);
+ } while (!zb->has_wp);
+
+ zone_lock(td, f, zb);
+ }
+
if (!zbd_open_zone(td, f, zb)) {
zone_unlock(zb);
zb = zbd_convert_to_open_zone(td, io_u);
}
}
+ if (zbd_zone_remainder(zb) > 0 &&
+ zbd_zone_remainder(zb) < min_bs)
+ goto retry;
+
/* Check whether the zone reset threshold has been exceeded */
if (td->o.zrf.u.f) {
if (zbdi->wp_sectors_with_data >= f->io_size * td->o.zrt.u.f &&
/* Reset the zone pointer if necessary */
if (zb->reset_zone || zbd_zone_full(f, zb, min_bs)) {
- assert(td->o.verify == VERIFY_NONE);
+ if (td->o.verify != VERIFY_NONE) {
+ /*
+ * Unset io-u->file to tell get_next_verify()
+ * that this IO is not requeue.
+ */
+ io_u->file = NULL;
+ if (!get_next_verify(td, io_u)) {
+ zone_unlock(zb);
+ return io_u_accept;
+ }
+ io_u->file = f;
+ }
+
/*
* Since previous write requests may have been submitted
* asynchronously and since we will submit the zone
/* Find out a non-empty zone to trim */
zone_unlock(zb);
- zl = get_zone(f, f->max_zone);
+ zl = zbd_get_zone(f, f->max_zone);
zb = zbd_find_zone(td, io_u, 1, zb, zl);
if (zb) {
io_u->offset = zb->start;
{
struct fio_file *f = io_u->file;
struct fio_zone_info *z;
- uint32_t zone_idx;
int ret;
- zone_idx = zbd_zone_idx(f, io_u->offset);
- z = get_zone(f, zone_idx);
-
+ z = zbd_offset_to_zone(f, io_u->offset);
if (!z->has_wp)
return 0;