drm/gm12u320: Replace system_long_wq with system_dfl_long_wq

[RFC PATCH] drm/gm12u320: Replace system_long_wq with system_dfl_long_wq

Posted by Marco Crivellari 1 month, 2 weeks ago

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/gpu/drm/tiny/gm12u320.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/tiny/gm12u320.c b/drivers/gpu/drm/tiny/gm12u320.c
index d73dfebb4353..4ad074337af0 100644
--- a/drivers/gpu/drm/tiny/gm12u320.c
+++ b/drivers/gpu/drm/tiny/gm12u320.c
@@ -388,7 +388,7 @@ static void gm12u320_fb_update_work(struct work_struct *work)
 	 * We must draw a frame every 2s otherwise the projector
 	 * switches back to showing its logo.
 	 */
-	queue_delayed_work(system_long_wq, &gm12u320->fb_update.work,
+	queue_delayed_work(system_dfl_long_wq, &gm12u320->fb_update.work,
 			   msecs_to_jiffies(IDLE_TIMEOUT));
 
 	return;
@@ -427,7 +427,8 @@ static void gm12u320_fb_mark_dirty(struct drm_framebuffer *fb,
 	mutex_unlock(&gm12u320->fb_update.lock);
 
 	if (wakeup)
-		mod_delayed_work(system_long_wq, &gm12u320->fb_update.work, 0);
+		mod_delayed_work(system_dfl_long_wq,
+				 &gm12u320->fb_update.work, 0);
 
 	if (old_fb)
 		drm_framebuffer_put(old_fb);
-- 
2.53.0

Re: [RFC PATCH] drm/gm12u320: Replace system_long_wq with system_dfl_long_wq

Posted by Marco Crivellari 1 week, 4 days ago

Hi,

On Thu, Apr 30, 2026 at 11:15 AM Marco Crivellari
<marco.crivellari@suse.com> wrote:
> [...]
>  drivers/gpu/drm/tiny/gm12u320.c | 5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)

Gentle ping.

Thanks!

-- 

Marco Crivellari

SUSE Labs

Re: [RFC PATCH] drm/gm12u320: Replace system_long_wq with system_dfl_long_wq

Posted by Thomas Zimmermann 1 week, 4 days ago

Hi

Am 05.06.26 um 10:17 schrieb Marco Crivellari:
> Hi,
>
> On Thu, Apr 30, 2026 at 11:15 AM Marco Crivellari
> <marco.crivellari@suse.com> wrote:
>> [...]
>>   drivers/gpu/drm/tiny/gm12u320.c | 5 +++--
>>   1 file changed, 3 insertions(+), 2 deletions(-)
> Gentle ping.

This patch as already been merged a bit ago.

https://gitlab.freedesktop.org/drm/misc/kernel/-/commit/645abe00a34588f27c856ca2b532642a8ef7691b

Best regards
Thomas

>
> Thanks!
>

-- 
--
Thomas Zimmermann
Graphics Driver Developer
SUSE Software Solutions Germany GmbH
Frankenstr. 146, 90461 Nürnberg, Germany, www.suse.com
GF: Jochen Jaser, Andrew McDonald, Werner Knoblich, (HRB 36809, AG Nürnberg)

Re: [RFC PATCH] drm/gm12u320: Replace system_long_wq with system_dfl_long_wq

Posted by Marco Crivellari 1 week, 4 days ago

On Fri, Jun 5, 2026 at 10:22 AM Thomas Zimmermann <tzimmermann@suse.de> wrote:
>
> Hi
>
> Am 05.06.26 um 10:17 schrieb Marco Crivellari:
> > Hi,
> >
> > On Thu, Apr 30, 2026 at 11:15 AM Marco Crivellari
> > <marco.crivellari@suse.com> wrote:
> >> [...]
> >>   drivers/gpu/drm/tiny/gm12u320.c | 5 +++--
> >>   1 file changed, 3 insertions(+), 2 deletions(-)
> > Gentle ping.
>
> This patch as already been merged a bit ago.

Many thanks Thomas!
-- 

Marco Crivellari

SUSE Labs

Re: [RFC PATCH] drm/gm12u320: Replace system_long_wq with system_dfl_long_wq

Posted by Thomas Zimmermann 1 month, 1 week ago


Am 30.04.26 um 11:15 schrieb Marco Crivellari:
> 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>

Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>

> ---
>   drivers/gpu/drm/tiny/gm12u320.c | 5 +++--
>   1 file changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/tiny/gm12u320.c b/drivers/gpu/drm/tiny/gm12u320.c
> index d73dfebb4353..4ad074337af0 100644
> --- a/drivers/gpu/drm/tiny/gm12u320.c
> +++ b/drivers/gpu/drm/tiny/gm12u320.c
> @@ -388,7 +388,7 @@ static void gm12u320_fb_update_work(struct work_struct *work)
>   	 * We must draw a frame every 2s otherwise the projector
>   	 * switches back to showing its logo.
>   	 */
> -	queue_delayed_work(system_long_wq, &gm12u320->fb_update.work,
> +	queue_delayed_work(system_dfl_long_wq, &gm12u320->fb_update.work,
>   			   msecs_to_jiffies(IDLE_TIMEOUT));
>   
>   	return;
> @@ -427,7 +427,8 @@ static void gm12u320_fb_mark_dirty(struct drm_framebuffer *fb,
>   	mutex_unlock(&gm12u320->fb_update.lock);
>   
>   	if (wakeup)
> -		mod_delayed_work(system_long_wq, &gm12u320->fb_update.work, 0);
> +		mod_delayed_work(system_dfl_long_wq,
> +				 &gm12u320->fb_update.work, 0);
>   
>   	if (old_fb)
>   		drm_framebuffer_put(old_fb);

-- 
--
Thomas Zimmermann
Graphics Driver Developer
SUSE Software Solutions Germany GmbH
Frankenstr. 146, 90461 Nürnberg, Germany, www.suse.com
GF: Jochen Jaser, Andrew McDonald, Werner Knoblich, (HRB 36809, AG Nürnberg)