* [PATCH] driver core: use READ_ONCE() for dev->driver in dev_has_sync_state()
@ 2026-04-18 16:22 Danilo Krummrich
2026-04-18 19:16 ` Saravana Kannan
0 siblings, 1 reply; 2+ messages in thread
From: Danilo Krummrich @ 2026-04-18 16:22 UTC (permalink / raw)
To: gregkh, rafael, saravanak, ulf.hansson
Cc: driver-core, linux-kernel, Danilo Krummrich
dev_has_sync_state() reads dev->driver twice without holding
device_lock() -- once for the NULL check and once to dereference
->sync_state. Some callers only hold device_links_write_lock, which
doesn't prevent a concurrent unbind from clearing dev->driver via
device_unbind_cleanup().
Fix it by reading dev->driver exactly once with READ_ONCE(), pairing
with the WRITE_ONCE() in device_set_driver().
Link: https://lore.kernel.org/driver-core/DHW8QPU1VU1F.3P6PH69HLFBYC@kernel.org/
Fixes: ac338acf514e ("driver core: Add dev_has_sync_state()")
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
include/linux/device.h | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/include/linux/device.h b/include/linux/device.h
index ac972e7bead4..4c1c9cb8570a 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -1018,9 +1018,12 @@ static inline void device_lock_assert(struct device *dev)
static inline bool dev_has_sync_state(struct device *dev)
{
+ struct device_driver *drv;
+
if (!dev)
return false;
- if (dev->driver && dev->driver->sync_state)
+ drv = READ_ONCE(dev->driver);
+ if (drv && drv->sync_state)
return true;
if (dev->bus && dev->bus->sync_state)
return true;
base-commit: 5b484311507b5d403c1f7a45f6aa3778549e268b
--
2.53.0
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH] driver core: use READ_ONCE() for dev->driver in dev_has_sync_state()
2026-04-18 16:22 [PATCH] driver core: use READ_ONCE() for dev->driver in dev_has_sync_state() Danilo Krummrich
@ 2026-04-18 19:16 ` Saravana Kannan
0 siblings, 0 replies; 2+ messages in thread
From: Saravana Kannan @ 2026-04-18 19:16 UTC (permalink / raw)
To: Danilo Krummrich
Cc: gregkh, rafael, saravanak, ulf.hansson, driver-core, linux-kernel
On Sat, Apr 18, 2026 at 9:22 AM Danilo Krummrich <dakr@kernel.org> wrote:
>
> dev_has_sync_state() reads dev->driver twice without holding
> device_lock() -- once for the NULL check and once to dereference
> ->sync_state. Some callers only hold device_links_write_lock, which
> doesn't prevent a concurrent unbind from clearing dev->driver via
> device_unbind_cleanup().
>
> Fix it by reading dev->driver exactly once with READ_ONCE(), pairing
> with the WRITE_ONCE() in device_set_driver().
>
> Link: https://lore.kernel.org/driver-core/DHW8QPU1VU1F.3P6PH69HLFBYC@kernel.org/
> Fixes: ac338acf514e ("driver core: Add dev_has_sync_state()")
> Signed-off-by: Danilo Krummrich <dakr@kernel.org>
Reviewed-by: Saravana Kannan <saravanak@kernel.org>
-Saravana
> ---
> include/linux/device.h | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/include/linux/device.h b/include/linux/device.h
> index ac972e7bead4..4c1c9cb8570a 100644
> --- a/include/linux/device.h
> +++ b/include/linux/device.h
> @@ -1018,9 +1018,12 @@ static inline void device_lock_assert(struct device *dev)
>
> static inline bool dev_has_sync_state(struct device *dev)
> {
> + struct device_driver *drv;
> +
> if (!dev)
> return false;
> - if (dev->driver && dev->driver->sync_state)
> + drv = READ_ONCE(dev->driver);
> + if (drv && drv->sync_state)
> return true;
> if (dev->bus && dev->bus->sync_state)
> return true;
>
> base-commit: 5b484311507b5d403c1f7a45f6aa3778549e268b
> --
> 2.53.0
>
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2026-04-18 19:16 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-04-18 16:22 [PATCH] driver core: use READ_ONCE() for dev->driver in dev_has_sync_state() Danilo Krummrich
2026-04-18 19:16 ` Saravana Kannan
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox