[PATCH v2] PM: sleep: wakeirq: harden dev_pm_clear_wake_irq() against races

[PATCH v2] PM: sleep: wakeirq: harden dev_pm_clear_wake_irq() against races

[Gui-Dong Han]

dev_pm_clear_wake_irq() currently uses a dangerous pattern where
dev->power.wakeirq is read and checked for NULL outside the lock. If two
callers invoke this function concurrently, both might see a valid
pointer and proceed. This could result in a double-free when the second
caller acquires the lock and tries to release the same object.

Address this by removing the lockless check of dev->power.wakeirq.
Instead, acquire dev->power.lock immediately to ensure the check and the
subsequent operations are atomic. If dev->power.wakeirq is NULL under
the lock, simply unlock and return. This guarantees that concurrent
calls cannot race to free the same object.

Based on a quick scan of current users, I did not find an actual bug as
drivers seem to rely on their own synchronization. However, since
asynchronous usage patterns exist (e.g., in
drivers/net/wireless/ti/wlcore), I believe a race is theoretically
possible if the API is used less carefully in the future. This change
hardens the API to be robust against such cases.

Fixes: 4990d4fe327b ("PM / Wakeirq: Add automated device wake IRQ handling")
Signed-off-by: Gui-Dong Han <hanguidong02@gmail.com>
---
@Rafael J. Wysocki: While studying wakeirq.c, I noticed comments for
dev_pm_enable_wake_irq_check() and friends are outdated. They claim
"Caller must hold &dev->power.lock" and limit usage to
rpm_suspend/resume, yet pm_runtime_force_suspend/resume() call them
lockless. Should I submit a follow-up patch to simply remove these
restrictions or complicate the text with exceptions?

v2:
* Remove the lockless check and perform the check protected by the lock
to avoid races, as suggested by Rafael J. Wysocki.
v1:
* Initial fix attempt using double-checked locking.
---
 drivers/base/power/wakeirq.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/base/power/wakeirq.c b/drivers/base/power/wakeirq.c
index 8aa28c08b289..c0809d18fc54 100644
--- a/drivers/base/power/wakeirq.c
+++ b/drivers/base/power/wakeirq.c
@@ -83,13 +83,16 @@ EXPORT_SYMBOL_GPL(dev_pm_set_wake_irq);
  */
 void dev_pm_clear_wake_irq(struct device *dev)
 {
-	struct wake_irq *wirq = dev->power.wakeirq;
+	struct wake_irq *wirq;
 	unsigned long flags;
 
-	if (!wirq)
+	spin_lock_irqsave(&dev->power.lock, flags);
+	wirq = dev->power.wakeirq;
+	if (!wirq) {
+		spin_unlock_irqrestore(&dev->power.lock, flags);
 		return;
+	}
 
-	spin_lock_irqsave(&dev->power.lock, flags);
 	device_wakeup_detach_irq(dev);
 	dev->power.wakeirq = NULL;
 	spin_unlock_irqrestore(&dev->power.lock, flags);
-- 
2.43.0

Re: [PATCH v2] PM: sleep: wakeirq: harden dev_pm_clear_wake_irq() against races

[Markus Elfring]

…
> +++ b/drivers/base/power/wakeirq.c
> @@ -83,13 +83,16 @@ EXPORT_SYMBOL_GPL(dev_pm_set_wake_irq);
>   */
>  void dev_pm_clear_wake_irq(struct device *dev)
>  {
> -	struct wake_irq *wirq = dev->power.wakeirq;
> +	struct wake_irq *wirq;
>  	unsigned long flags;
>  
> -	if (!wirq)
> +	spin_lock_irqsave(&dev->power.lock, flags);
> +	wirq = dev->power.wakeirq;
> +	if (!wirq) {
> +		spin_unlock_irqrestore(&dev->power.lock, flags);
>  		return;
> +	}
>  
> -	spin_lock_irqsave(&dev->power.lock, flags);
>  	device_wakeup_detach_irq(dev);
>  	dev->power.wakeirq = NULL;
>  	spin_unlock_irqrestore(&dev->power.lock, flags);

Would it become feasible to apply a scoped_guard() call?
https://elixir.bootlin.com/linux/v6.19-rc5/source/include/linux/spinlock.h#L585

Regards,
Markus

Re: [PATCH v2] PM: sleep: wakeirq: harden dev_pm_clear_wake_irq() against races

[Rafael J. Wysocki]

On Tue, Feb 3, 2026 at 10:16 AM Markus Elfring <Markus.Elfring@web.de> wrote:
>
> …
> > +++ b/drivers/base/power/wakeirq.c
> > @@ -83,13 +83,16 @@ EXPORT_SYMBOL_GPL(dev_pm_set_wake_irq);
> >   */
> >  void dev_pm_clear_wake_irq(struct device *dev)
> >  {
> > -     struct wake_irq *wirq = dev->power.wakeirq;
> > +     struct wake_irq *wirq;
> >       unsigned long flags;
> >
> > -     if (!wirq)
> > +     spin_lock_irqsave(&dev->power.lock, flags);
> > +     wirq = dev->power.wakeirq;
> > +     if (!wirq) {
> > +             spin_unlock_irqrestore(&dev->power.lock, flags);
> >               return;
> > +     }
> >
> > -     spin_lock_irqsave(&dev->power.lock, flags);
> >       device_wakeup_detach_irq(dev);
> >       dev->power.wakeirq = NULL;
> >       spin_unlock_irqrestore(&dev->power.lock, flags);
>
> Would it become feasible to apply a scoped_guard() call?
> https://elixir.bootlin.com/linux/v6.19-rc5/source/include/linux/spinlock.h#L585

Yes, it would, but in a separate patch.

Re: [PATCH v2] PM: sleep: wakeirq: harden dev_pm_clear_wake_irq() against races

[Rafael J. Wysocki]

On Tue, Feb 3, 2026 at 4:20 AM Gui-Dong Han <hanguidong02@gmail.com> wrote:
>
> dev_pm_clear_wake_irq() currently uses a dangerous pattern where
> dev->power.wakeirq is read and checked for NULL outside the lock. If two
> callers invoke this function concurrently, both might see a valid
> pointer and proceed. This could result in a double-free when the second
> caller acquires the lock and tries to release the same object.
>
> Address this by removing the lockless check of dev->power.wakeirq.
> Instead, acquire dev->power.lock immediately to ensure the check and the
> subsequent operations are atomic. If dev->power.wakeirq is NULL under
> the lock, simply unlock and return. This guarantees that concurrent
> calls cannot race to free the same object.
>
> Based on a quick scan of current users, I did not find an actual bug as
> drivers seem to rely on their own synchronization. However, since
> asynchronous usage patterns exist (e.g., in
> drivers/net/wireless/ti/wlcore), I believe a race is theoretically
> possible if the API is used less carefully in the future. This change
> hardens the API to be robust against such cases.
>
> Fixes: 4990d4fe327b ("PM / Wakeirq: Add automated device wake IRQ handling")
> Signed-off-by: Gui-Dong Han <hanguidong02@gmail.com>

Patch applied as 6.20/7.0 material.

> ---
> @Rafael J. Wysocki: While studying wakeirq.c, I noticed comments for
> dev_pm_enable_wake_irq_check() and friends are outdated. They claim
> "Caller must hold &dev->power.lock" and limit usage to
> rpm_suspend/resume, yet pm_runtime_force_suspend/resume() call them
> lockless. Should I submit a follow-up patch to simply remove these
> restrictions or complicate the text with exceptions?

Please just send a separate patch to fix the outdated comments.

Thanks!

> v2:
> * Remove the lockless check and perform the check protected by the lock
> to avoid races, as suggested by Rafael J. Wysocki.
> v1:
> * Initial fix attempt using double-checked locking.
> ---
>  drivers/base/power/wakeirq.c | 9 ++++++---
>  1 file changed, 6 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/base/power/wakeirq.c b/drivers/base/power/wakeirq.c
> index 8aa28c08b289..c0809d18fc54 100644
> --- a/drivers/base/power/wakeirq.c
> +++ b/drivers/base/power/wakeirq.c
> @@ -83,13 +83,16 @@ EXPORT_SYMBOL_GPL(dev_pm_set_wake_irq);
>   */
>  void dev_pm_clear_wake_irq(struct device *dev)
>  {
> -       struct wake_irq *wirq = dev->power.wakeirq;
> +       struct wake_irq *wirq;
>         unsigned long flags;
>
> -       if (!wirq)
> +       spin_lock_irqsave(&dev->power.lock, flags);
> +       wirq = dev->power.wakeirq;
> +       if (!wirq) {
> +               spin_unlock_irqrestore(&dev->power.lock, flags);
>                 return;
> +       }
>
> -       spin_lock_irqsave(&dev->power.lock, flags);
>         device_wakeup_detach_irq(dev);
>         dev->power.wakeirq = NULL;
>         spin_unlock_irqrestore(&dev->power.lock, flags);
> --
> 2.43.0
>
>