On Thu, Jun 13, 2024 at 02:14:48PM +0200, Javier Carrasco wrote:
> The device_for_each_child_node() macro requires explicit calls to
> fwnode_handle_put() in all early exits of the loop if the child node is
> not required outside. Otherwise, the child node's refcount is not
> decremented and the resource is not released.
>
> The current implementation of pmic_glink_ucsi_probe() makes use of the
> device_for_each_child_node(), but does not release the child node on
> early returns. Add the missing calls to fwnode_handle_put().
>
> Cc: stable@xxxxxxxxxxxxxxx
> Fixes: c6165ed2f425 ("usb: ucsi: glink: use the connector orientation GPIO to provide switch events")
> Signed-off-by: Javier Carrasco <javier.carrasco.cruz@xxxxxxxxx>
> ---
> This case would be a great opportunity for the recently introduced
> device_for_each_child_node_scoped(), but given that it has not been
> released yet, the traditional approach has been used to account for
> stable kernels (bug introduced with v6.7). A second patch to clean
> this up with that macro is ready to be sent once this fix is applied,
> so this kind of problem does not arise if more early returns are added.
>
> This issue has been found while analyzing the code and not tested with
> hardware, only compiled and checked with static analysis tools. Any
> tests with real hardware are always welcome.
> ---
> drivers/usb/typec/ucsi/ucsi_glink.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@xxxxxxxxxx>
--
With best wishes
Dmitry
