run_fio "${opts[@]}" >> "${logfile}.${test_number}" 2>&1 || return $?
}
-# Verify that zone_reset_threshold only takes logical blocks from seq
-# zones into account, and logical blocks of conv zones are not counted.
+# Verify that zone_reset_threshold only accounts written bytes in seq
+# zones, and written data bytes of conv zones are not counted.
test52() {
local off io_size
grep -q 'not support experimental verify' "${logfile}.${test_number}"
}
+# Test fio errors out zone_reset_threshold option for multiple jobs with
+# different write ranges.
+test61() {
+ run_fio_on_seq "$(ioengine "psync")" --rw=write --size="$zone_size" \
+ --numjobs=2 --offset_increment="$zone_size" \
+ --zone_reset_threshold=0.1 --zone_reset_frequency=1 \
+ --exitall_on_error=1 \
+ >> "${logfile}.${test_number}" 2>&1 && return 1
+ grep -q 'different write ranges' "${logfile}.${test_number}"
+}
+
+# Test zone_reset_threshold option works for multiple jobs with same write
+# range.
+test62() {
+ local bs loops=2 size=$((zone_size))
+
+ [ -n "$is_zbd" ] && reset_zone "$dev" -1
+
+ # Two jobs write to single zone twice. Reset zone happens at next write
+ # after half of the zone gets filled. So 2 * 2 * 2 - 1 = 7 times zone
+ # resets are expected.
+ bs=$(min $((256*1024)) $((zone_size / 4)))
+ run_fio_on_seq "$(ioengine "psync")" --rw=write --bs="$bs" \
+ --size=$size --loops=$loops --numjobs=2 \
+ --zone_reset_frequency=1 --zone_reset_threshold=.5 \
+ --group_reporting=1 \
+ >> "${logfile}.${test_number}" 2>&1 || return $?
+ check_written $((size * loops * 2)) || return $?
+ check_reset_count -eq 7 || return $?
+}
+
+# Test zone_reset_threshold option works for a read job and a write job with
+# different IO range.
+test63() {
+ local bs loops=2 size=$((zone_size)) off1 off2
+
+ [ -n "$is_zbd" ] && reset_zone "$dev" -1
+
+ off1=$((first_sequential_zone_sector * 512))
+ off2=$((off1 + zone_size))
+ bs=$(min $((256*1024)) $((zone_size / 4)))
+
+ # One job writes to single zone twice. Reset zone happens at next write
+ # after half of the zone gets filled. So 2 * 2 - 1 = 3 times zone resets
+ # are expected.
+ run_fio "$(ioengine "psync")" --bs="$bs" --size=$size --loops=$loops \
+ --filename="$dev" --group_reporting=1 \
+ --zonemode=zbd --zonesize="$zone_size" --direct=1 \
+ --zone_reset_frequency=1 --zone_reset_threshold=.5 \
+ --name=r --rw=read --offset=$off1 "${job_var_opts[@]}" \
+ --name=w --rw=write --offset=$off2 "${job_var_opts[@]}" \
+ >> "${logfile}.${test_number}" 2>&1 || return $?
+ check_written $((size * loops)) || return $?
+ check_reset_count -eq 3 || return $?
+}
+
SECONDS=0
tests=()
dynamic_analyzer=()
return zbd_get_zone(f, zbd_offset_to_zone_idx(f, offset));
}
+static bool accounting_vdb(struct thread_data *td, const struct fio_file *f)
+{
+ return td->o.zrt.u.f && td_write(td);
+}
+
/**
* zbd_get_zoned_model - Get a device zoned model
* @td: FIO thread data
break;
}
- 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);
+ if (accounting_vdb(td, f)) {
+ pthread_mutex_lock(&f->zbd_info->mutex);
+ f->zbd_info->wp_valid_data_bytes -= data_in_zone;
+ pthread_mutex_unlock(&f->zbd_info->mutex);
+ }
z->wp = z->start;
}
/* Whether or not the I/O range for f includes one or more sequential zones */
-static bool zbd_is_seq_job(struct fio_file *f)
+static bool zbd_is_seq_job(const struct fio_file *f)
{
uint32_t zone_idx, zone_idx_b, zone_idx_e;
}
}
+static uint64_t zbd_verify_and_set_vdb(struct thread_data *td,
+ const struct fio_file *f)
+{
+ struct fio_zone_info *zb, *ze, *z;
+ uint64_t wp_vdb = 0;
+ struct zoned_block_device_info *zbdi = f->zbd_info;
+
+ assert(td->runstate < TD_RUNNING);
+ assert(zbdi);
+
+ if (!accounting_vdb(td, f))
+ return 0;
+
+ /*
+ * Ensure that the I/O range includes one or more sequential zones so
+ * that f->min_zone and f->max_zone have different values.
+ */
+ if (!zbd_is_seq_job(f))
+ return 0;
+
+ if (zbdi->write_min_zone != zbdi->write_max_zone) {
+ if (zbdi->write_min_zone != f->min_zone ||
+ zbdi->write_max_zone != f->max_zone) {
+ td_verror(td, EINVAL,
+ "multi-jobs with different write ranges are "
+ "not supported with zone_reset_threshold");
+ log_err("multi-jobs with different write ranges are "
+ "not supported with zone_reset_threshold\n");
+ }
+ return 0;
+ }
+
+ zbdi->write_min_zone = f->min_zone;
+ zbdi->write_max_zone = 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)
+ wp_vdb += z->wp - z->start;
+
+ zbdi->wp_valid_data_bytes = wp_vdb;
+
+ return wp_vdb;
+}
+
int zbd_setup_files(struct thread_data *td)
{
struct fio_file *f;
struct zoned_block_device_info *zbd = f->zbd_info;
struct fio_zone_info *z;
int zi;
+ uint64_t vdb;
assert(zbd);
f->max_zone =
zbd_offset_to_zone_idx(f, f->file_offset + f->io_size);
+ vdb = zbd_verify_and_set_vdb(td, f);
+
+ dprint(FD_ZBD, "%s(%s): valid data bytes = %" PRIu64 "\n",
+ __func__, f->file_name, vdb);
+
/*
* When all zones in the I/O range are conventional, io_size
* can be smaller than zone size, making min_zone the same
return write_cnt == 0;
}
-enum swd_action {
- CHECK_SWD,
- SET_SWD,
-};
-
-/* Calculate the number of sectors with data (swd) and perform action 'a' */
-static uint64_t zbd_process_swd(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 = 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);
- 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++)
- if (z->has_wp)
- zone_unlock(z);
-
- return swd;
-}
-
-/*
- * The swd check is useful for debugging but takes too much time to leave
- * it enabled all the time. Hence it is disabled by default.
- */
-static const bool enable_check_swd = false;
-
-/* Check whether the 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(td, f, CHECK_SWD);
-}
-
void zbd_file_reset(struct thread_data *td, struct fio_file *f)
{
struct fio_zone_info *zb, *ze;
- uint64_t swd;
bool verify_data_left = false;
if (!f->zbd_info || !td_write(td))
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",
- __func__, f->file_name, swd);
/*
* If data verification is enabled reset the affected zones before
* z->wp > zone_end means that one or more I/O errors
* have occurred.
*/
- pthread_mutex_lock(&zbd_info->mutex);
- if (z->wp <= zone_end) {
- zbd_info->sectors_with_data += zone_end - z->wp;
- zbd_info->wp_sectors_with_data += zone_end - z->wp;
+ if (accounting_vdb(td, f) && z->wp <= zone_end) {
+ pthread_mutex_lock(&zbd_info->mutex);
+ zbd_info->wp_valid_data_bytes += zone_end - z->wp;
+ pthread_mutex_unlock(&zbd_info->mutex);
}
- pthread_mutex_unlock(&zbd_info->mutex);
z->wp = zone_end;
break;
default:
zbd_end_zone_io(td, io_u, z);
zone_unlock(z);
- zbd_check_swd(td, f);
}
/*
if (ddir != DDIR_READ || !td_rw(td))
return ddir;
- if (io_u->file->zbd_info->sectors_with_data ||
- td->o.read_beyond_wp)
+ if (io_u->file->last_start[DDIR_WRITE] != -1ULL || td->o.read_beyond_wp)
return DDIR_READ;
return DDIR_WRITE;
io_u->ddir == DDIR_READ && td->o.read_beyond_wp)
return io_u_accept;
- zbd_check_swd(td, f);
-
zone_lock(td, f, zb);
switch (io_u->ddir) {
/* 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 &&
+ if (zbdi->wp_valid_data_bytes >=
+ f->io_size * td->o.zrt.u.f &&
zbd_dec_and_reset_write_cnt(td, f))
zb->reset_zone = 1;
}