[PATCH] ata: libata-sff: add WQ_PERCPU to alloc_workqueue users

[PATCH] ata: libata-sff: add WQ_PERCPU to alloc_workqueue users

[Marco Crivellari]

Currently if a user enqueue 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.

alloc_workqueue() treats all queues as per-CPU by default, while unbound
workqueues must opt-in via WQ_UNBOUND.

This default is suboptimal: most workloads benefit from unbound queues,
allowing the scheduler to place worker threads where they’re needed and
reducing noise when CPUs are isolated.

This continues the effort to refactor workqueue APIs, which began with
the introduction of new workqueues and a new alloc_workqueue flag in:

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

This change adds a new WQ_PERCPU flag to explicitly request
alloc_workqueue() to be per-cpu when WQ_UNBOUND has not been specified.

With the introduction of the WQ_PERCPU flag (equivalent to !WQ_UNBOUND),
any alloc_workqueue() caller that doesn’t explicitly specify WQ_UNBOUND
must now use WQ_PERCPU.

Once migration is complete, WQ_UNBOUND can be removed and unbound will
become the implicit default.

Suggested-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
---
 drivers/ata/libata-sff.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 1e2a2c33cdc8..785b6e371abf 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -3191,7 +3191,8 @@ void ata_sff_port_init(struct ata_port *ap)
 
 int __init ata_sff_init(void)
 {
-	ata_sff_wq = alloc_workqueue("ata_sff", WQ_MEM_RECLAIM, WQ_MAX_ACTIVE);
+	ata_sff_wq = alloc_workqueue("ata_sff", WQ_MEM_RECLAIM | WQ_PERCPU,
+				     WQ_MAX_ACTIVE);
 	if (!ata_sff_wq)
 		return -ENOMEM;
 
-- 
2.51.1

Re: [PATCH] ata: libata-sff: add WQ_PERCPU to alloc_workqueue users

[Damien Le Moal]

On 11/7/25 01:08, Marco Crivellari wrote:
> Currently if a user enqueue 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.
> 
> alloc_workqueue() treats all queues as per-CPU by default, while unbound
> workqueues must opt-in via WQ_UNBOUND.
> 
> This default is suboptimal: most workloads benefit from unbound queues,
> allowing the scheduler to place worker threads where they’re needed and
> reducing noise when CPUs are isolated.
> 
> This continues the effort to refactor workqueue APIs, which began with
> the introduction of new workqueues and a new alloc_workqueue flag in:
> 
> commit 128ea9f6ccfb ("workqueue: Add system_percpu_wq and system_dfl_wq")
> commit 930c2ea566af ("workqueue: Add new WQ_PERCPU flag")
> 
> This change adds a new WQ_PERCPU flag to explicitly request
> alloc_workqueue() to be per-cpu when WQ_UNBOUND has not been specified.
> 
> With the introduction of the WQ_PERCPU flag (equivalent to !WQ_UNBOUND),
> any alloc_workqueue() caller that doesn’t explicitly specify WQ_UNBOUND
> must now use WQ_PERCPU.
> 
> Once migration is complete, WQ_UNBOUND can be removed and unbound will
> become the implicit default.
> 
> Suggested-by: Tejun Heo <tj@kernel.org>
> Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>

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


-- 
Damien Le Moal
Western Digital Research

Re: [PATCH] ata: libata-sff: add WQ_PERCPU to alloc_workqueue users

[Niklas Cassel]

On Thu, 06 Nov 2025 17:08:13 +0100, Marco Crivellari wrote:
> Currently if a user enqueue 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.
> 
> [...]

Applied to libata/linux.git (for-6.19), thanks!

[1/1] ata: libata-sff: add WQ_PERCPU to alloc_workqueue users
      https://git.kernel.org/libata/linux/c/13f4d995

Kind regards,
Niklas

Re: [PATCH] ata: libata-sff: add WQ_PERCPU to alloc_workqueue users

[Marco Crivellari]

On Fri, Nov 7, 2025 at 9:45 AM Niklas Cassel <cassel@kernel.org> wrote:
>[...]
> Applied to libata/linux.git (for-6.19), thanks!
>
> [1/1] ata: libata-sff: add WQ_PERCPU to alloc_workqueue users
>       https://git.kernel.org/libata/linux/c/13f4d995
>
> Kind regards,
> Niklas
>

Many thanks!

-- 

Marco Crivellari

L3 Support Engineer, Technology & Product