[PATCH] nvdimm: replace use of system_wq with system_percpu_wq

[PATCH] nvdimm: replace use of system_wq with system_percpu_wq

[Marco Crivellari]

Currently if a user enqueues a work item using schedule_delayed_work() the
used wq is "system_wq" (per-cpu wq) while queue_delayed_work() use
WORK_CPU_UNBOUND (used when a cpu is not specified). The same applies to
schedule_work() that is using system_wq and queue_work(), that makes use
again of WORK_CPU_UNBOUND.

This lack of consistency cannot be addressed without refactoring the API.

This patch continues the effort to refactor worqueue APIs, which has begun
with the change introducing new workqueues and a new alloc_workqueue flag:

commit 128ea9f6ccfb ("workqueue: Add system_percpu_wq and system_dfl_wq")
commit 930c2ea566af ("workqueue: Add new WQ_PERCPU flag")

Replace system_wq with system_percpu_wq, keeping the same old behavior.
The old wq (system_wq) will be kept for a few release cycles.

Suggested-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
---
 drivers/nvdimm/security.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/nvdimm/security.c b/drivers/nvdimm/security.c
index 4adce8c38870..e41f6951ca0f 100644
--- a/drivers/nvdimm/security.c
+++ b/drivers/nvdimm/security.c
@@ -424,7 +424,7 @@ static int security_overwrite(struct nvdimm *nvdimm, unsigned int keyid)
 		 * query.
 		 */
 		get_device(dev);
-		queue_delayed_work(system_wq, &nvdimm->dwork, 0);
+		queue_delayed_work(system_percpu_wq, &nvdimm->dwork, 0);
 	}
 
 	return rc;
@@ -457,7 +457,7 @@ static void __nvdimm_security_overwrite_query(struct nvdimm *nvdimm)
 
 		/* setup delayed work again */
 		tmo += 10;
-		queue_delayed_work(system_wq, &nvdimm->dwork, tmo * HZ);
+		queue_delayed_work(system_percpu_wq, &nvdimm->dwork, tmo * HZ);
 		nvdimm->sec.overwrite_tmo = min(15U * 60U, tmo);
 		return;
 	}
-- 
2.51.1

Re: [PATCH] nvdimm: replace use of system_wq with system_percpu_wq

[Dave Jiang]

On 11/5/25 8:08 AM, Marco Crivellari wrote:
> Currently if a user enqueues a work item using schedule_delayed_work() the
> used wq is "system_wq" (per-cpu wq) while queue_delayed_work() use
> WORK_CPU_UNBOUND (used when a cpu is not specified). The same applies to
> schedule_work() that is using system_wq and queue_work(), that makes use
> again of WORK_CPU_UNBOUND.
> 
> This lack of consistency cannot be addressed without refactoring the API.
> 
> This patch continues the effort to refactor worqueue APIs, which has begun
> with the change introducing new workqueues and a new alloc_workqueue flag:
> 
> commit 128ea9f6ccfb ("workqueue: Add system_percpu_wq and system_dfl_wq")
> commit 930c2ea566af ("workqueue: Add new WQ_PERCPU flag")
> 
> Replace system_wq with system_percpu_wq, keeping the same old behavior.
> The old wq (system_wq) will be kept for a few release cycles.
> 
> Suggested-by: Tejun Heo <tj@kernel.org>
> Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>> ---
>  drivers/nvdimm/security.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/nvdimm/security.c b/drivers/nvdimm/security.c
> index 4adce8c38870..e41f6951ca0f 100644
> --- a/drivers/nvdimm/security.c
> +++ b/drivers/nvdimm/security.c
> @@ -424,7 +424,7 @@ static int security_overwrite(struct nvdimm *nvdimm, unsigned int keyid)
>  		 * query.
>  		 */
>  		get_device(dev);
> -		queue_delayed_work(system_wq, &nvdimm->dwork, 0);
> +		queue_delayed_work(system_percpu_wq, &nvdimm->dwork, 0);
>  	}
>  
>  	return rc;
> @@ -457,7 +457,7 @@ static void __nvdimm_security_overwrite_query(struct nvdimm *nvdimm)
>  
>  		/* setup delayed work again */
>  		tmo += 10;
> -		queue_delayed_work(system_wq, &nvdimm->dwork, tmo * HZ);
> +		queue_delayed_work(system_percpu_wq, &nvdimm->dwork, tmo * HZ);
>  		nvdimm->sec.overwrite_tmo = min(15U * 60U, tmo);
>  		return;
>  	}

Re: [PATCH] nvdimm: replace use of system_wq with system_percpu_wq

[Ira Weiny]

Marco Crivellari wrote:
> Currently if a user enqueues a work item using schedule_delayed_work() the
> used wq is "system_wq" (per-cpu wq) while queue_delayed_work() use
> WORK_CPU_UNBOUND (used when a cpu is not specified). The same applies to
> schedule_work() that is using system_wq and queue_work(), that makes use
> again of WORK_CPU_UNBOUND.
> 
> This lack of consistency cannot be addressed without refactoring the API.
> 
> This patch continues the effort to refactor worqueue APIs, which has begun
> with the change introducing new workqueues and a new alloc_workqueue flag:
> 
> commit 128ea9f6ccfb ("workqueue: Add system_percpu_wq and system_dfl_wq")
> commit 930c2ea566af ("workqueue: Add new WQ_PERCPU flag")
> 
> Replace system_wq with system_percpu_wq, keeping the same old behavior.
> The old wq (system_wq) will be kept for a few release cycles.
> 
> Suggested-by: Tejun Heo <tj@kernel.org>
> Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>

Queued Thanks,

Reviewed-by: Ira Weiny <ira.weiny@intel.com>

[snip]

Re: [PATCH] nvdimm: replace use of system_wq with system_percpu_wq

[Marco Crivellari]

On Wed, Nov 5, 2025 at 10:46 PM Ira Weiny <ira.weiny@intel.com> wrote:
>[...]
> Queued Thanks,
>
> Reviewed-by: Ira Weiny <ira.weiny@intel.com>

Many thanks!

-- 

Marco Crivellari

L3 Support Engineer, Technology & Product