driver_set_override() modifies and frees dev->driver_override while
holding device_lock(dev). However, driver_match_device() reads
dev->driver_override when calling bus match functions.

Currently, driver_match_device() is called from three sites. One site
(__device_attach_driver) holds device_lock(dev), but the other two
(bind_store and __driver_attach) do not. This allows a concurrent
driver_set_override() to free the string while driver_match_device() is
using it, leading to a use-after-free (UAF).

This issue affects at least 11 bus types (including PCI, AMBA, Platform)
that rely on driver_override for matching.

Fix this by holding device_lock(dev) around the driver_match_device() calls
in bind_store() and __driver_attach(). This ensures all access to
dev->driver_override via driver_match_device() is protected by the device
lock. Also add a lock assertion to driver_match_device() to prevent future
locking regressions.

Tested with the PoCs from Bugzilla that trigger this UAF. Stress testing
the two newly locked paths for 24 hours with CONFIG_PROVE_LOCKING and
CONFIG_LOCKDEP enabled showed no UAF recurrence and no lockdep
warnings.

Closes: https://bugzilla.kernel.org/show_bug.cgi?id=220789
Suggested-by: Qiu-ji Chen <chenqiuji666@gmail.com>
Signed-off-by: Gui-Dong Han <hanguidong02@gmail.com>
---
v2:
* Add device_lock_assert() in driver_match_device() to enforce locking
requirement, as suggested by Greg KH.
v1:
* The Bugzilla entry contains full KASAN reports and two PoCs that reliably
reproduce the UAF on both unlocked paths using a standard QEMU setup
(default e1000 device at 0000:00:03.0).
I chose to fix this in the driver core for the following reasons:
1. Both racing functions are part of the driver core.
2. Fixing this per-driver/per-bus is tedious and would require careful
ad-hoc locking that does not align with the existing device_lock(dev).
3. We cannot simply add device_lock(dev) inside bus match functions because
one call path (__device_attach_driver) already holds this lock. Adding the
lock inside the match callback would cause a deadlock on that path.
---
 drivers/base/base.h |  2 ++
 drivers/base/bus.c  | 17 ++++++++++++-----
 drivers/base/dd.c   |  3 +++
 3 files changed, 17 insertions(+), 5 deletions(-)

diff --git a/drivers/base/base.h b/drivers/base/base.h
index 86fa7fbb3548..011e910a53f8 100644
--- a/drivers/base/base.h
+++ b/drivers/base/base.h
@@ -166,6 +166,8 @@ void device_set_deferred_probe_reason(const struct device *dev, struct va_format
 static inline int driver_match_device(const struct device_driver *drv,
 				      struct device *dev)
 {
+	/* Protects against driver_set_override() races */
+	device_lock_assert(dev);
 	return drv->bus->match ? drv->bus->match(dev, drv) : 1;
 }
 
diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index 5e75e1bce551..9e62d6009058 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -261,13 +261,20 @@ static ssize_t bind_store(struct device_driver *drv, const char *buf,
 	const struct bus_type *bus = bus_get(drv->bus);
 	struct device *dev;
 	int err = -ENODEV;
+	int ret;
 
 	dev = bus_find_device_by_name(bus, NULL, buf);
-	if (dev && driver_match_device(drv, dev)) {
-		err = device_driver_attach(drv, dev);
-		if (!err) {
-			/* success */
-			err = count;
+	if (dev) {
+		/* Protects against driver_set_override() races */
+		device_lock(dev);
+		ret = driver_match_device(drv, dev);
+		device_unlock(dev);
+		if (ret) {
+			err = device_driver_attach(drv, dev);
+			if (!err) {
+				/* success */
+				err = count;
+			}
 		}
 	}
 	put_device(dev);
diff --git a/drivers/base/dd.c b/drivers/base/dd.c
index 13ab98e033ea..db60b4500136 100644
--- a/drivers/base/dd.c
+++ b/drivers/base/dd.c
@@ -1170,7 +1170,10 @@ static int __driver_attach(struct device *dev, void *data)
 	 * is an error.
 	 */
 
+	/* Protects against driver_set_override() races */
+	device_lock(dev);
 	ret = driver_match_device(drv, dev);
+	device_unlock(dev);
 	if (ret == 0) {
 		/* no match */
 		return 0;
-- 
2.43.0

On Thu, Nov 27, 2025 at 12:00:11AM +0800, Gui-Dong Han wrote:
> driver_set_override() modifies and frees dev->driver_override while
> holding device_lock(dev). However, driver_match_device() reads
> dev->driver_override when calling bus match functions.
> 
> Currently, driver_match_device() is called from three sites. One site
> (__device_attach_driver) holds device_lock(dev), but the other two
> (bind_store and __driver_attach) do not. This allows a concurrent
> driver_set_override() to free the string while driver_match_device() is
> using it, leading to a use-after-free (UAF).
> 
> This issue affects at least 11 bus types (including PCI, AMBA, Platform)
> that rely on driver_override for matching.
> 
> Fix this by holding device_lock(dev) around the driver_match_device() calls
> in bind_store() and __driver_attach(). This ensures all access to
> dev->driver_override via driver_match_device() is protected by the device
> lock. Also add a lock assertion to driver_match_device() to prevent future
> locking regressions.
> 
> Tested with the PoCs from Bugzilla that trigger this UAF. Stress testing
> the two newly locked paths for 24 hours with CONFIG_PROVE_LOCKING and
> CONFIG_LOCKDEP enabled showed no UAF recurrence and no lockdep
> warnings.
> 
> Closes: https://bugzilla.kernel.org/show_bug.cgi?id=220789
> Suggested-by: Qiu-ji Chen <chenqiuji666@gmail.com>
> Signed-off-by: Gui-Dong Han <hanguidong02@gmail.com>
> ---
> v2:
> * Add device_lock_assert() in driver_match_device() to enforce locking
> requirement, as suggested by Greg KH.
> v1:
> * The Bugzilla entry contains full KASAN reports and two PoCs that reliably
> reproduce the UAF on both unlocked paths using a standard QEMU setup
> (default e1000 device at 0000:00:03.0).
> I chose to fix this in the driver core for the following reasons:
> 1. Both racing functions are part of the driver core.
> 2. Fixing this per-driver/per-bus is tedious and would require careful
> ad-hoc locking that does not align with the existing device_lock(dev).
> 3. We cannot simply add device_lock(dev) inside bus match functions because
> one call path (__device_attach_driver) already holds this lock. Adding the
> lock inside the match callback would cause a deadlock on that path.
> ---
>  drivers/base/base.h |  2 ++
>  drivers/base/bus.c  | 17 ++++++++++++-----
>  drivers/base/dd.c   |  3 +++
>  3 files changed, 17 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/base/base.h b/drivers/base/base.h
> index 86fa7fbb3548..011e910a53f8 100644
> --- a/drivers/base/base.h
> +++ b/drivers/base/base.h
> @@ -166,6 +166,8 @@ void device_set_deferred_probe_reason(const struct device *dev, struct va_format
>  static inline int driver_match_device(const struct device_driver *drv,
>  				      struct device *dev)
>  {
> +	/* Protects against driver_set_override() races */
> +	device_lock_assert(dev);
>  	return drv->bus->match ? drv->bus->match(dev, drv) : 1;

A new line after the assert?

And thanks for adding the comment, but:

>  }
>  
> diff --git a/drivers/base/bus.c b/drivers/base/bus.c
> index 5e75e1bce551..9e62d6009058 100644
> --- a/drivers/base/bus.c
> +++ b/drivers/base/bus.c
> @@ -261,13 +261,20 @@ static ssize_t bind_store(struct device_driver *drv, const char *buf,
>  	const struct bus_type *bus = bus_get(drv->bus);
>  	struct device *dev;
>  	int err = -ENODEV;
> +	int ret;
>  
>  	dev = bus_find_device_by_name(bus, NULL, buf);
> -	if (dev && driver_match_device(drv, dev)) {
> -		err = device_driver_attach(drv, dev);
> -		if (!err) {
> -			/* success */
> -			err = count;
> +	if (dev) {
> +		/* Protects against driver_set_override() races */

This comment isn't needed anymore.

> +		device_lock(dev);
> +		ret = driver_match_device(drv, dev);
> +		device_unlock(dev);
> +		if (ret) {
> +			err = device_driver_attach(drv, dev);
> +			if (!err) {
> +				/* success */
> +				err = count;
> +			}
>  		}
>  	}
>  	put_device(dev);
> diff --git a/drivers/base/dd.c b/drivers/base/dd.c
> index 13ab98e033ea..db60b4500136 100644
> --- a/drivers/base/dd.c
> +++ b/drivers/base/dd.c
> @@ -1170,7 +1170,10 @@ static int __driver_attach(struct device *dev, void *data)
>  	 * is an error.
>  	 */
>  
> +	/* Protects against driver_set_override() races */

Nor is this one.

Also, I looked again, and why are you not grabbing the lock in the
bind_store() sysfs call?  driver_match_device() is only called in 3
places, 2 now have the lock held, and one does not.  This feels wrong.

thanks,

greg k-h

On Thu, Nov 27, 2025 at 5:16 PM Greg KH <gregkh@linuxfoundation.org> wrote:
>
> On Thu, Nov 27, 2025 at 12:00:11AM +0800, Gui-Dong Han wrote:
> > driver_set_override() modifies and frees dev->driver_override while
> > holding device_lock(dev). However, driver_match_device() reads
> > dev->driver_override when calling bus match functions.
> >
> > Currently, driver_match_device() is called from three sites. One site
> > (__device_attach_driver) holds device_lock(dev), but the other two
> > (bind_store and __driver_attach) do not. This allows a concurrent
> > driver_set_override() to free the string while driver_match_device() is
> > using it, leading to a use-after-free (UAF).
> >
> > This issue affects at least 11 bus types (including PCI, AMBA, Platform)
> > that rely on driver_override for matching.
> >
> > Fix this by holding device_lock(dev) around the driver_match_device() calls
> > in bind_store() and __driver_attach(). This ensures all access to
> > dev->driver_override via driver_match_device() is protected by the device
> > lock. Also add a lock assertion to driver_match_device() to prevent future
> > locking regressions.
> >
> > Tested with the PoCs from Bugzilla that trigger this UAF. Stress testing
> > the two newly locked paths for 24 hours with CONFIG_PROVE_LOCKING and
> > CONFIG_LOCKDEP enabled showed no UAF recurrence and no lockdep
> > warnings.
> >
> > Closes: https://bugzilla.kernel.org/show_bug.cgi?id=220789
> > Suggested-by: Qiu-ji Chen <chenqiuji666@gmail.com>
> > Signed-off-by: Gui-Dong Han <hanguidong02@gmail.com>
> > ---
> > v2:
> > * Add device_lock_assert() in driver_match_device() to enforce locking
> > requirement, as suggested by Greg KH.
> > v1:
> > * The Bugzilla entry contains full KASAN reports and two PoCs that reliably
> > reproduce the UAF on both unlocked paths using a standard QEMU setup
> > (default e1000 device at 0000:00:03.0).
> > I chose to fix this in the driver core for the following reasons:
> > 1. Both racing functions are part of the driver core.
> > 2. Fixing this per-driver/per-bus is tedious and would require careful
> > ad-hoc locking that does not align with the existing device_lock(dev).
> > 3. We cannot simply add device_lock(dev) inside bus match functions because
> > one call path (__device_attach_driver) already holds this lock. Adding the
> > lock inside the match callback would cause a deadlock on that path.
> > ---
> >  drivers/base/base.h |  2 ++
> >  drivers/base/bus.c  | 17 ++++++++++++-----
> >  drivers/base/dd.c   |  3 +++
> >  3 files changed, 17 insertions(+), 5 deletions(-)
> >
> > diff --git a/drivers/base/base.h b/drivers/base/base.h
> > index 86fa7fbb3548..011e910a53f8 100644
> > --- a/drivers/base/base.h
> > +++ b/drivers/base/base.h
> > @@ -166,6 +166,8 @@ void device_set_deferred_probe_reason(const struct device *dev, struct va_format
> >  static inline int driver_match_device(const struct device_driver *drv,
> >                                     struct device *dev)
> >  {
> > +     /* Protects against driver_set_override() races */
> > +     device_lock_assert(dev);
> >       return drv->bus->match ? drv->bus->match(dev, drv) : 1;
>
> A new line after the assert?
>
> And thanks for adding the comment, but:
>
> >  }
> >
> > diff --git a/drivers/base/bus.c b/drivers/base/bus.c
> > index 5e75e1bce551..9e62d6009058 100644
> > --- a/drivers/base/bus.c
> > +++ b/drivers/base/bus.c
> > @@ -261,13 +261,20 @@ static ssize_t bind_store(struct device_driver *drv, const char *buf,
> >       const struct bus_type *bus = bus_get(drv->bus);
> >       struct device *dev;
> >       int err = -ENODEV;
> > +     int ret;
> >
> >       dev = bus_find_device_by_name(bus, NULL, buf);
> > -     if (dev && driver_match_device(drv, dev)) {
> > -             err = device_driver_attach(drv, dev);
> > -             if (!err) {
> > -                     /* success */
> > -                     err = count;
> > +     if (dev) {
> > +             /* Protects against driver_set_override() races */
>
> This comment isn't needed anymore.
>
> > +             device_lock(dev);
> > +             ret = driver_match_device(drv, dev);
> > +             device_unlock(dev);
> > +             if (ret) {
> > +                     err = device_driver_attach(drv, dev);
> > +                     if (!err) {
> > +                             /* success */
> > +                             err = count;
> > +                     }
> >               }
> >       }
> >       put_device(dev);
> > diff --git a/drivers/base/dd.c b/drivers/base/dd.c
> > index 13ab98e033ea..db60b4500136 100644
> > --- a/drivers/base/dd.c
> > +++ b/drivers/base/dd.c
> > @@ -1170,7 +1170,10 @@ static int __driver_attach(struct device *dev, void *data)
> >        * is an error.
> >        */
> >
> > +     /* Protects against driver_set_override() races */
>
> Nor is this one.
>
> Also, I looked again, and why are you not grabbing the lock in the
> bind_store() sysfs call?  driver_match_device() is only called in 3
> places, 2 now have the lock held, and one does not.  This feels wrong.

Actually, the change in drivers/base/bus.c is exactly for the
bind_store() path. So all 3 call sites are covered and locked with
this patch.

I have verified this with the PoC for bind_store(), and the new
lockdep assertion does not trigger any warnings, confirming the lock
is properly held.

I have just sent a v3 patch which adds the blank line and removes the
redundant comments as suggested.

Thanks,
Gui-Dong Han

On Thu, Nov 27, 2025 at 11:00:16PM +0800, Gui-Dong Han wrote:
> On Thu, Nov 27, 2025 at 5:16 PM Greg KH <gregkh@linuxfoundation.org> wrote:
> >
> > On Thu, Nov 27, 2025 at 12:00:11AM +0800, Gui-Dong Han wrote:
> > > driver_set_override() modifies and frees dev->driver_override while
> > > holding device_lock(dev). However, driver_match_device() reads
> > > dev->driver_override when calling bus match functions.
> > >
> > > Currently, driver_match_device() is called from three sites. One site
> > > (__device_attach_driver) holds device_lock(dev), but the other two
> > > (bind_store and __driver_attach) do not. This allows a concurrent
> > > driver_set_override() to free the string while driver_match_device() is
> > > using it, leading to a use-after-free (UAF).
> > >
> > > This issue affects at least 11 bus types (including PCI, AMBA, Platform)
> > > that rely on driver_override for matching.
> > >
> > > Fix this by holding device_lock(dev) around the driver_match_device() calls
> > > in bind_store() and __driver_attach(). This ensures all access to
> > > dev->driver_override via driver_match_device() is protected by the device
> > > lock. Also add a lock assertion to driver_match_device() to prevent future
> > > locking regressions.
> > >
> > > Tested with the PoCs from Bugzilla that trigger this UAF. Stress testing
> > > the two newly locked paths for 24 hours with CONFIG_PROVE_LOCKING and
> > > CONFIG_LOCKDEP enabled showed no UAF recurrence and no lockdep
> > > warnings.
> > >
> > > Closes: https://bugzilla.kernel.org/show_bug.cgi?id=220789
> > > Suggested-by: Qiu-ji Chen <chenqiuji666@gmail.com>
> > > Signed-off-by: Gui-Dong Han <hanguidong02@gmail.com>
> > > ---
> > > v2:
> > > * Add device_lock_assert() in driver_match_device() to enforce locking
> > > requirement, as suggested by Greg KH.
> > > v1:
> > > * The Bugzilla entry contains full KASAN reports and two PoCs that reliably
> > > reproduce the UAF on both unlocked paths using a standard QEMU setup
> > > (default e1000 device at 0000:00:03.0).
> > > I chose to fix this in the driver core for the following reasons:
> > > 1. Both racing functions are part of the driver core.
> > > 2. Fixing this per-driver/per-bus is tedious and would require careful
> > > ad-hoc locking that does not align with the existing device_lock(dev).
> > > 3. We cannot simply add device_lock(dev) inside bus match functions because
> > > one call path (__device_attach_driver) already holds this lock. Adding the
> > > lock inside the match callback would cause a deadlock on that path.
> > > ---
> > >  drivers/base/base.h |  2 ++
> > >  drivers/base/bus.c  | 17 ++++++++++++-----
> > >  drivers/base/dd.c   |  3 +++
> > >  3 files changed, 17 insertions(+), 5 deletions(-)
> > >
> > > diff --git a/drivers/base/base.h b/drivers/base/base.h
> > > index 86fa7fbb3548..011e910a53f8 100644
> > > --- a/drivers/base/base.h
> > > +++ b/drivers/base/base.h
> > > @@ -166,6 +166,8 @@ void device_set_deferred_probe_reason(const struct device *dev, struct va_format
> > >  static inline int driver_match_device(const struct device_driver *drv,
> > >                                     struct device *dev)
> > >  {
> > > +     /* Protects against driver_set_override() races */
> > > +     device_lock_assert(dev);
> > >       return drv->bus->match ? drv->bus->match(dev, drv) : 1;
> >
> > A new line after the assert?
> >
> > And thanks for adding the comment, but:
> >
> > >  }
> > >
> > > diff --git a/drivers/base/bus.c b/drivers/base/bus.c
> > > index 5e75e1bce551..9e62d6009058 100644
> > > --- a/drivers/base/bus.c
> > > +++ b/drivers/base/bus.c
> > > @@ -261,13 +261,20 @@ static ssize_t bind_store(struct device_driver *drv, const char *buf,
> > >       const struct bus_type *bus = bus_get(drv->bus);
> > >       struct device *dev;
> > >       int err = -ENODEV;
> > > +     int ret;
> > >
> > >       dev = bus_find_device_by_name(bus, NULL, buf);
> > > -     if (dev && driver_match_device(drv, dev)) {
> > > -             err = device_driver_attach(drv, dev);
> > > -             if (!err) {
> > > -                     /* success */
> > > -                     err = count;
> > > +     if (dev) {
> > > +             /* Protects against driver_set_override() races */
> >
> > This comment isn't needed anymore.
> >
> > > +             device_lock(dev);
> > > +             ret = driver_match_device(drv, dev);
> > > +             device_unlock(dev);
> > > +             if (ret) {
> > > +                     err = device_driver_attach(drv, dev);
> > > +                     if (!err) {
> > > +                             /* success */
> > > +                             err = count;
> > > +                     }
> > >               }
> > >       }
> > >       put_device(dev);
> > > diff --git a/drivers/base/dd.c b/drivers/base/dd.c
> > > index 13ab98e033ea..db60b4500136 100644
> > > --- a/drivers/base/dd.c
> > > +++ b/drivers/base/dd.c
> > > @@ -1170,7 +1170,10 @@ static int __driver_attach(struct device *dev, void *data)
> > >        * is an error.
> > >        */
> > >
> > > +     /* Protects against driver_set_override() races */
> >
> > Nor is this one.
> >
> > Also, I looked again, and why are you not grabbing the lock in the
> > bind_store() sysfs call?  driver_match_device() is only called in 3
> > places, 2 now have the lock held, and one does not.  This feels wrong.
> 
> Actually, the change in drivers/base/bus.c is exactly for the
> bind_store() path. So all 3 call sites are covered and locked with
> this patch.

Ugh, nevermind, your patch changes bind_store() I missed that, my fault
:(

> I have verified this with the PoC for bind_store(), and the new
> lockdep assertion does not trigger any warnings, confirming the lock
> is properly held.

Makes sense, thanks!

I'll queue up your patch after 6.19-rc1 is out, thanks!

greg k-h