[PATCH v2 0/2] zbd: fix open zone status check for non-write jobs

[PATCH v2 0/2] zbd: fix open zone status check for non-write jobs

[Shin'ichiro Kawasaki]

Currently, fio checks if each zone within the IO range is in an open
condition or not. This check is done even if the job does not perform
any write operations, and causes confusion among users.

This series addresses the problem. The first patch fixes it, and the
second patch adds a test case to confirm the fix.

Changes from v1:
* 1st patch: added Fixes tags

Shin'ichiro Kawasaki (2):
  zbd: do not check open zones status and limits when jobs do not write
  t/zbd: add test case to confirm no max_open_zones limit check

 t/zbd/test-zbd-support | 29 +++++++++++++++++++++++++++++
 zbd.c                  | 10 ++++++++++
 2 files changed, 39 insertions(+)

-- 
2.47.0

[PATCH v2 1/2] zbd: do not check open zones status and limits when jobs do not write

[Shin'ichiro Kawasaki]

Currently, fio checks the conditions of each zone within the IO range at
the job start. If a zone is in an open condition, it is added to the
write target zone array. If the number of write target zones exceeds the
max_open_zones or job_max_open_zones limit, fio terminates with the
error message "Number of open zones exceeds max_open_zones limit". This
check for zone condition and the resulting termination occur even when
the job does not perform a write operation, leading to confusion among
users.

To avoid the confusion, skip the check when jobs do not perform write
operations. Additionally, print the message to inform that the
job_max_open_zones limit does not work for non-write jobs.

Fixes: 954217b90191 ("zbd: Initialize open zones list referring zone status at fio start")
Fixes: 8ac768899d63 ("zbd: do not reset extra zones in open conditions")
Signed-off-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
---
 zbd.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/zbd.c b/zbd.c
index 8a092cbe..ee095b1d 100644
--- a/zbd.c
+++ b/zbd.c
@@ -1264,6 +1264,16 @@ int zbd_setup_files(struct thread_data *td)
 			return 1;
 		}
 
+		/*
+		 * If this job does not do write operations, skip open zone
+		 * condition check.
+		 */
+		if (!td_write(td)) {
+			if (td->o.job_max_open_zones)
+				log_info("'job_max_open_zones' is valid only for write jobs\n");
+			continue;
+		}
+
 		/*
 		 * The per job max open zones limit cannot be used without a
 		 * global max open zones limit. (As the tracking of open zones
-- 
2.47.0

[PATCH v2 2/2] t/zbd: add test case to confirm no max_open_zones limit check

[Shin'ichiro Kawasaki]

The previous commit fixed the max_open_zones limit check for non-write
jobs. Add a test case to confirm the fix.

Signed-off-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
---
 t/zbd/test-zbd-support | 29 +++++++++++++++++++++++++++++
 1 file changed, 29 insertions(+)

diff --git a/t/zbd/test-zbd-support b/t/zbd/test-zbd-support
index e0b2a755..468fce70 100755
--- a/t/zbd/test-zbd-support
+++ b/t/zbd/test-zbd-support
@@ -1608,6 +1608,35 @@ test69() {
 		>> "${logfile}.${test_number}" 2>&1 || return $?
 }
 
+# Test max_open_zones and job_max_open_zones do not error out for non-write jobs
+test70() {
+	require_zbd || return "$SKIP_TESTCASE"
+
+	reset_zone "${dev}" -1
+
+	# Write data to two zones and make them open
+	run_fio_on_seq "$(ioengine "psync")" --io_size="$min_seq_write_size" \
+		       --rw=write --offset_increment=1z --numjobs=2 \
+		       --group_reporting=1 >> "${logfile}.${test_number}" 2>&1
+
+	# Confirm max_open_zones=1 for read workload does not fail
+	run_fio_on_seq "$(ioengine "psync")" --io_size="$min_seq_write_size" \
+		       --rw=read --max_open_zones=1 \
+		       >> "${logfile}.${test_number}" 2>&1 || return $?
+
+	# Confirm job_max_open_zones=1 for read workload does not fail
+	run_fio_on_seq "$(ioengine "psync")" --io_size="$min_seq_write_size" \
+		       --rw=read --job_max_open_zones=1 \
+		       >> "${logfile}.${test_number}" 2>&1
+	grep -q 'valid only for write jobs' \
+	     "${logfile}.${test_number}" || return $?
+
+	# Confirm max_open_zones=1 for trim workload does not fail
+	run_fio_on_seq "$(ioengine "psync")" --rw=trim --io_size=1z \
+		       --bs="$zone_size" --max_open_zones=1 \
+		       >> "${logfile}.${test_number}" 2>&1
+}
+
 SECONDS=0
 tests=()
 dynamic_analyzer=()
-- 
2.47.0

Re: [PATCH v2 0/2] zbd: fix open zone status check for non-write jobs

[fiotestbot]

[-- Attachment #1: Type: text/plain, Size: 144 bytes --]


The result of fio's continuous integration tests was: success

For more details see https://github.com/fiotestbot/fio/actions/runs/12899960142

Re: [PATCH v2 0/2] zbd: fix open zone status check for non-write jobs

[Damien Le Moal]

On 1/22/25 10:38 AM, Shin'ichiro Kawasaki wrote:
> Currently, fio checks if each zone within the IO range is in an open
> condition or not. This check is done even if the job does not perform
> any write operations, and causes confusion among users.
> 
> This series addresses the problem. The first patch fixes it, and the
> second patch adds a test case to confirm the fix.

Looks good to me. For the series:

Reviewed-by: Damien Le Moal <dlemoal@kernel.org>

-- 
Damien Le Moal
Western Digital Research

Re: [PATCH v2 1/2] zbd: do not check open zones status and limits when jobs do not write

[Niklas Cassel]

On Wed, Jan 22, 2025 at 10:38:34AM +0900, Shin'ichiro Kawasaki wrote:
> Currently, fio checks the conditions of each zone within the IO range at
> the job start. If a zone is in an open condition, it is added to the
> write target zone array. If the number of write target zones exceeds the
> max_open_zones or job_max_open_zones limit, fio terminates with the
> error message "Number of open zones exceeds max_open_zones limit". This
> check for zone condition and the resulting termination occur even when
> the job does not perform a write operation, leading to confusion among
> users.
> 
> To avoid the confusion, skip the check when jobs do not perform write
> operations. Additionally, print the message to inform that the
> job_max_open_zones limit does not work for non-write jobs.
> 
> Fixes: 954217b90191 ("zbd: Initialize open zones list referring zone status at fio start")
> Fixes: 8ac768899d63 ("zbd: do not reset extra zones in open conditions")
> Signed-off-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
> ---
>  zbd.c | 10 ++++++++++
>  1 file changed, 10 insertions(+)
> 
> diff --git a/zbd.c b/zbd.c
> index 8a092cbe..ee095b1d 100644
> --- a/zbd.c
> +++ b/zbd.c
> @@ -1264,6 +1264,16 @@ int zbd_setup_files(struct thread_data *td)
>  			return 1;
>  		}
>  
> +		/*
> +		 * If this job does not do write operations, skip open zone
> +		 * condition check.
> +		 */
> +		if (!td_write(td)) {
> +			if (td->o.job_max_open_zones)
> +				log_info("'job_max_open_zones' is valid only for write jobs\n");
> +			continue;
> +		}
> +
>  		/*
>  		 * The per job max open zones limit cannot be used without a
>  		 * global max open zones limit. (As the tracking of open zones
> -- 
> 2.47.0
> 

Reviewed-by: Niklas Cassel <niklas.cassel@wdc.com>

Re: [PATCH v2 0/2] zbd: fix open zone status check for non-write jobs

[Vincent Fu]

On 1/21/25 8:38 PM, Shin'ichiro Kawasaki wrote:
> Currently, fio checks if each zone within the IO range is in an open
> condition or not. This check is done even if the job does not perform
> any write operations, and causes confusion among users.
> 
> This series addresses the problem. The first patch fixes it, and the
> second patch adds a test case to confirm the fix.
> 
> Changes from v1:
> * 1st patch: added Fixes tags
> 
> Shin'ichiro Kawasaki (2):
>    zbd: do not check open zones status and limits when jobs do not write
>    t/zbd: add test case to confirm no max_open_zones limit check
> 
>   t/zbd/test-zbd-support | 29 +++++++++++++++++++++++++++++
>   zbd.c                  | 10 ++++++++++
>   2 files changed, 39 insertions(+)
> 


Applied. Thanks.

Vincent