Re: [PATCH] regulator: core: Fix error return for /sys access

[Adam Ford]

On Tue, Sep 10, 2019 at 2:22 PM H. Nikolaus Schaller <hns@xxxxxxxxxxxxx> wrote:
>
> regulator_uV_show() is missing error handling if regulator_get_voltage_rdev()
> returns negative values. Instead it prints the errno as a string, e.g. -EINVAL
> as "-22" which could be interpreted as -22 ÂV.
>
> We also do not need to hold the lock while converting the integer to a string.
>
> Reported-by: Adam Ford <aford173@xxxxxxxxx>
> Signed-off-by: H. Nikolaus Schaller <hns@xxxxxxxxxxxxx>

Tested-by: Adam Ford <aford173@xxxxxxxxx>

> ---
>  drivers/regulator/core.c | 8 +++++---
>  1 file changed, 5 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c
> index e0c0cf462004..ea48cb5a68b8 100644
> --- a/drivers/regulator/core.c
> +++ b/drivers/regulator/core.c
> @@ -564,13 +564,15 @@ static ssize_t regulator_uV_show(struct device *dev,
>                                 struct device_attribute *attr, char *buf)
>  {
>         struct regulator_dev *rdev = dev_get_drvdata(dev);
> -       ssize_t ret;
> +       int uV;
>
>         regulator_lock(rdev);
> -       ret = sprintf(buf, "%d\n", regulator_get_voltage_rdev(rdev));
> +       uV = regulator_get_voltage_rdev(rdev);
>         regulator_unlock(rdev);
>
> -       return ret;
> +       if (uV < 0)
> +               return uV;
> +       return sprintf(buf, "%d\n", uV);
>  }
>  static DEVICE_ATTR(microvolts, 0444, regulator_uV_show, NULL);
>
> --
> 2.19.1
>