[PATCH v1] cpuidle: Fail cpuidle device registration if there is one already

Posted by Rafael J. Wysocki 1 week, 6 days ago

From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

Refuse to register a cpuidle device if the given CPU has a cpuidle
device already and print a message regarding it.

Without this, an attempt to register a new cpuidle device without
unregistering the existing one leads to the removal of the existing
cpuidle device without removing its sysfs interface.

Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
---
 drivers/cpuidle/cpuidle.c |    8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

--- a/drivers/cpuidle/cpuidle.c
+++ b/drivers/cpuidle/cpuidle.c
@@ -635,11 +635,17 @@ static void __cpuidle_device_init(struct
 static int __cpuidle_register_device(struct cpuidle_device *dev)
 {
 	struct cpuidle_driver *drv = cpuidle_get_cpu_driver(dev);
+	unsigned int cpu = dev->cpu;
 	int i, ret;
 
 	if (!try_module_get(drv->owner))
 		return -EINVAL;
 
+	if (per_cpu(cpuidle_devices, cpu)) {
+		pr_info("CPU%d: cpuidle device already registered\n", cpu);
+		return -EEXIST;
+	}
+
 	for (i = 0; i < drv->state_count; i++) {
 		if (drv->states[i].flags & CPUIDLE_FLAG_UNUSABLE)
 			dev->states_usage[i].disable |= CPUIDLE_STATE_DISABLED_BY_DRIVER;
@@ -648,7 +654,7 @@ static int __cpuidle_register_device(str
 			dev->states_usage[i].disable |= CPUIDLE_STATE_DISABLED_BY_USER;
 	}
 
-	per_cpu(cpuidle_devices, dev->cpu) = dev;
+	per_cpu(cpuidle_devices, cpu) = dev;
 	list_add(&dev->device_list, &cpuidle_detected_devices);
 
 	ret = cpuidle_coupled_register_device(dev);

Re: [PATCH v1] cpuidle: Fail cpuidle device registration if there is one already

Posted by Dhruva Gole 1 week, 6 days ago

Hi Rafael,

On Sep 18, 2025 at 23:19:20 +0200, Rafael J. Wysocki wrote:
> From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> 
> Refuse to register a cpuidle device if the given CPU has a cpuidle
> device already and print a message regarding it.
> 
> Without this, an attempt to register a new cpuidle device without
> unregistering the existing one leads to the removal of the existing
> cpuidle device without removing its sysfs interface.
> 
> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> ---
>  drivers/cpuidle/cpuidle.c |    8 +++++++-
>  1 file changed, 7 insertions(+), 1 deletion(-)
> 
> --- a/drivers/cpuidle/cpuidle.c
> +++ b/drivers/cpuidle/cpuidle.c
> @@ -635,11 +635,17 @@ static void __cpuidle_device_init(struct
>  static int __cpuidle_register_device(struct cpuidle_device *dev)
>  {
>  	struct cpuidle_driver *drv = cpuidle_get_cpu_driver(dev);
> +	unsigned int cpu = dev->cpu;
>  	int i, ret;
>  
>  	if (!try_module_get(drv->owner))
>  		return -EINVAL;
>  
> +	if (per_cpu(cpuidle_devices, cpu)) {
> +		pr_info("CPU%d: cpuidle device already registered\n", cpu);
> +		return -EEXIST;

Here we return prematurely after a try_module_get right?
Do we need a module_put() similar to how you do it later by calling
unregister_device function by checking ret = cpuidle_coupled_register_device ?

> +	}
> +
>  	for (i = 0; i < drv->state_count; i++) {
>  		if (drv->states[i].flags & CPUIDLE_FLAG_UNUSABLE)
>  			dev->states_usage[i].disable |= CPUIDLE_STATE_DISABLED_BY_DRIVER;
> @@ -648,7 +654,7 @@ static int __cpuidle_register_device(str
>  			dev->states_usage[i].disable |= CPUIDLE_STATE_DISABLED_BY_USER;
>  	}
>  
> -	per_cpu(cpuidle_devices, dev->cpu) = dev;
> +	per_cpu(cpuidle_devices, cpu) = dev;
>  	list_add(&dev->device_list, &cpuidle_detected_devices);
>  
>  	ret = cpuidle_coupled_register_device(dev);
> 
> 
> 
> 

-- 
Best regards,
Dhruva Gole
Texas Instruments Incorporated

Re: [PATCH v1] cpuidle: Fail cpuidle device registration if there is one already

Posted by Rafael J. Wysocki 1 week, 5 days ago

Hi,

On Fri, Sep 19, 2025 at 7:09 AM Dhruva Gole <d-gole@ti.com> wrote:
>
> Hi Rafael,
>
> On Sep 18, 2025 at 23:19:20 +0200, Rafael J. Wysocki wrote:
> > From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> >
> > Refuse to register a cpuidle device if the given CPU has a cpuidle
> > device already and print a message regarding it.
> >
> > Without this, an attempt to register a new cpuidle device without
> > unregistering the existing one leads to the removal of the existing
> > cpuidle device without removing its sysfs interface.
> >
> > Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> > ---
> >  drivers/cpuidle/cpuidle.c |    8 +++++++-
> >  1 file changed, 7 insertions(+), 1 deletion(-)
> >
> > --- a/drivers/cpuidle/cpuidle.c
> > +++ b/drivers/cpuidle/cpuidle.c
> > @@ -635,11 +635,17 @@ static void __cpuidle_device_init(struct
> >  static int __cpuidle_register_device(struct cpuidle_device *dev)
> >  {
> >       struct cpuidle_driver *drv = cpuidle_get_cpu_driver(dev);
> > +     unsigned int cpu = dev->cpu;
> >       int i, ret;
> >
> >       if (!try_module_get(drv->owner))
> >               return -EINVAL;
> >
> > +     if (per_cpu(cpuidle_devices, cpu)) {
> > +             pr_info("CPU%d: cpuidle device already registered\n", cpu);
> > +             return -EEXIST;
>
> Here we return prematurely after a try_module_get right?
> Do we need a module_put() similar to how you do it later by calling
> unregister_device function by checking ret = cpuidle_coupled_register_device ?

Good catch, thanks!

I need to move the new check above the driver module reference counting.