drivers/i2c/i2c-slave-testunit.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
Currently the code enqueue work items using {queue|mod}_delayed_work(),
using system_long_wq. This workqueue should be used when long works are
expected, but it is a per-cpu workqueue.
This is important because queue_delayed_work() queue the work using:
queue_delayed_work_on(WORK_CPU_UNBOUND, ...);
Note that WORK_CPU_UNBOUND = NR_CPUS.
This would end up calling __queue_delayed_work() that does:
if (housekeeping_enabled(HK_TYPE_TIMER)) {
// [....]
} else {
if (likely(cpu == WORK_CPU_UNBOUND))
add_timer_global(timer);
else
add_timer_on(timer, cpu);
}
So when cpu == WORK_CPU_UNBOUND the timer is global and is
not using a specific CPU. Later, when __queue_work() is called:
if (req_cpu == WORK_CPU_UNBOUND) {
if (wq->flags & WQ_UNBOUND)
cpu = wq_select_unbound_cpu(raw_smp_processor_id());
else
cpu = raw_smp_processor_id();
}
Because the wq is not unbound, it takes the CPU where the timer
fired and enqueue the work on that CPU.
The consequence of all of this is that the work can run anywhere,
depending on where the timer fired.
Recently, a new unbound workqueue specific for long running work has
been added:
c116737e972e ("workqueue: Add system_dfl_long_wq for long unbound works")
So change system_long_wq with system_dfl_long_wq so that the work may
benefit from scheduler task placement.
Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
---
drivers/i2c/i2c-slave-testunit.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/i2c/i2c-slave-testunit.c b/drivers/i2c/i2c-slave-testunit.c
index 6de4307050dd..c6582ca86c7d 100644
--- a/drivers/i2c/i2c-slave-testunit.c
+++ b/drivers/i2c/i2c-slave-testunit.c
@@ -124,7 +124,7 @@ static int i2c_slave_testunit_slave_cb(struct i2c_client *client,
case I2C_SLAVE_STOP:
if (tu->reg_idx == TU_NUM_REGS) {
set_bit(TU_FLAG_IN_PROCESS, &tu->flags);
- queue_delayed_work(system_long_wq, &tu->worker,
+ queue_delayed_work(system_dfl_long_wq, &tu->worker,
msecs_to_jiffies(10 * tu->regs[TU_REG_DELAY]));
}
--
2.53.0On Thu, Apr 30, 2026 at 11:08:10AM +0200, Marco Crivellari wrote:
> Currently the code enqueue work items using {queue|mod}_delayed_work(),
> using system_long_wq. This workqueue should be used when long works are
> expected, but it is a per-cpu workqueue.
>
> This is important because queue_delayed_work() queue the work using:
>
> queue_delayed_work_on(WORK_CPU_UNBOUND, ...);
>
> Note that WORK_CPU_UNBOUND = NR_CPUS.
>
> This would end up calling __queue_delayed_work() that does:
>
> if (housekeeping_enabled(HK_TYPE_TIMER)) {
> // [....]
> } else {
> if (likely(cpu == WORK_CPU_UNBOUND))
> add_timer_global(timer);
> else
> add_timer_on(timer, cpu);
> }
>
> So when cpu == WORK_CPU_UNBOUND the timer is global and is
> not using a specific CPU. Later, when __queue_work() is called:
>
> if (req_cpu == WORK_CPU_UNBOUND) {
> if (wq->flags & WQ_UNBOUND)
> cpu = wq_select_unbound_cpu(raw_smp_processor_id());
> else
> cpu = raw_smp_processor_id();
> }
>
> Because the wq is not unbound, it takes the CPU where the timer
> fired and enqueue the work on that CPU.
> The consequence of all of this is that the work can run anywhere,
> depending on where the timer fired.
>
> Recently, a new unbound workqueue specific for long running work has
> been added:
>
> c116737e972e ("workqueue: Add system_dfl_long_wq for long unbound works")
>
> So change system_long_wq with system_dfl_long_wq so that the work may
> benefit from scheduler task placement.
>
> Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
Looks very reasonable to me. With your detailed explanation, we could
probably also remove the FIXME next to the workqueue.h-include. This
seems to be the proper workqueue now. I will fix it when applying.
Thank you!
On Thu, Apr 30, 2026 at 11:08:10AM +0200, Marco Crivellari wrote:
> Currently the code enqueue work items using {queue|mod}_delayed_work(),
> using system_long_wq. This workqueue should be used when long works are
> expected, but it is a per-cpu workqueue.
>
> This is important because queue_delayed_work() queue the work using:
>
> queue_delayed_work_on(WORK_CPU_UNBOUND, ...);
>
> Note that WORK_CPU_UNBOUND = NR_CPUS.
>
> This would end up calling __queue_delayed_work() that does:
>
> if (housekeeping_enabled(HK_TYPE_TIMER)) {
> // [....]
> } else {
> if (likely(cpu == WORK_CPU_UNBOUND))
> add_timer_global(timer);
> else
> add_timer_on(timer, cpu);
> }
>
> So when cpu == WORK_CPU_UNBOUND the timer is global and is
> not using a specific CPU. Later, when __queue_work() is called:
>
> if (req_cpu == WORK_CPU_UNBOUND) {
> if (wq->flags & WQ_UNBOUND)
> cpu = wq_select_unbound_cpu(raw_smp_processor_id());
> else
> cpu = raw_smp_processor_id();
> }
>
> Because the wq is not unbound, it takes the CPU where the timer
> fired and enqueue the work on that CPU.
> The consequence of all of this is that the work can run anywhere,
> depending on where the timer fired.
>
> Recently, a new unbound workqueue specific for long running work has
> been added:
>
> c116737e972e ("workqueue: Add system_dfl_long_wq for long unbound works")
>
> So change system_long_wq with system_dfl_long_wq so that the work may
> benefit from scheduler task placement.
>
> Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
Applied to for-current, thanks!
BTW, what does _dfl_ stand for?
On Mon, May 4, 2026 at 10:07 AM Wolfram Sang <wsa+renesas@sang-engineering.com> wrote: > [..] > Applied to for-current, thanks! Many thanks, Wolfram! I completely missed the FIXME on top. > > BTW, what does _dfl_ stand for? > "dfl" stands for "default". We're keeping the same naming convention used for the unbound workqueue, system_dfl_wq. Thanks! -- Marco Crivellari SUSE Labs
© 2016 - 2026 Red Hat, Inc.