On Fri, Jan 16, 2026 at 9:11 AM Tzung-Bi Shih <[email protected]> wrote:
>
> On error handling paths, gpiolib_cdev_register() doesn't free the
> allocated resources which results leaks. Fix it.
>
> Cc: [email protected]
> Fixes: 7b9b77a8bba9 ("gpiolib: add a per-gpio_device line state notification
> workqueue")
> Fixes: d83cee3d2bb1 ("gpio: protect the pointer to gpio_chip in gpio_device
> with SRCU")
> Signed-off-by: Tzung-Bi Shih <[email protected]>
> ---
> drivers/gpio/gpiolib-cdev.c | 13 ++++++++++---
> 1 file changed, 10 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpio/gpiolib-cdev.c b/drivers/gpio/gpiolib-cdev.c
> index 3735c9fe1502..ba1eae15852d 100644
> --- a/drivers/gpio/gpiolib-cdev.c
> +++ b/drivers/gpio/gpiolib-cdev.c
> @@ -2797,16 +2797,23 @@ int gpiolib_cdev_register(struct gpio_device *gdev,
> dev_t devt)
>
> ret = cdev_device_add(&gdev->chrdev, &gdev->dev);
> if (ret)
> - return ret;
> + goto err_free_workqueue;
>
I need to drop this because it jumps over the guard(). I think you'll
have to free the workqueue locally here instead.
Can you send a separate v2?
Bart
> guard(srcu)(&gdev->srcu);
> gc = srcu_dereference(gdev->chip, &gdev->srcu);
> - if (!gc)
> - return -ENODEV;
> + if (!gc) {
> + ret = -ENODEV;
> + goto err_free_cdev;
> + }
>
> gpiochip_dbg(gc, "added GPIO chardev (%d:%d)\n", MAJOR(devt),
> gdev->id);
>
> return 0;
> +err_free_cdev:
> + cdev_device_del(&gdev->chrdev, &gdev->dev);
> +err_free_workqueue:
> + destroy_workqueue(gdev->line_state_wq);
> + return ret;
> }
>
> void gpiolib_cdev_unregister(struct gpio_device *gdev)
> --
> 2.52.0.457.g6b5491de43-goog
>
