Re: [PATCH v1 1/1] backlight: lm3630a_bl: Put fwnode in error case during ->probe()

[Daniel Thompson]

On Mon, May 10, 2021 at 12:57:16PM +0300, Andy Shevchenko wrote:
> device_for_each_child_node() bumps a reference counting of a returned variable.
> We have to balance it whenever we return to the caller.
> 
> Fixes: 8fbce8efe15cd ("backlight: lm3630a: Add firmware node support")
> Cc: Brian Masney <masneyb@xxxxxxxxxxxxx>
> Cc: Dan Murphy <dmurphy@xxxxxx>
> Signed-off-by: Andy Shevchenko <andy.shevchenko@xxxxxxxxx>

Reviewed-by: Daniel Thompson <daniel.thompson@xxxxxxxxxx>


> ---
>  drivers/video/backlight/lm3630a_bl.c | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/video/backlight/lm3630a_bl.c b/drivers/video/backlight/lm3630a_bl.c
> index e88a2b0e5904..662029d6a3dc 100644
> --- a/drivers/video/backlight/lm3630a_bl.c
> +++ b/drivers/video/backlight/lm3630a_bl.c
> @@ -482,8 +482,10 @@ static int lm3630a_parse_node(struct lm3630a_chip *pchip,
>  
>  	device_for_each_child_node(pchip->dev, node) {
>  		ret = lm3630a_parse_bank(pdata, node, &seen_led_sources);
> -		if (ret)
> +		if (ret) {
> +			fwnode_handle_put(node);
>  			return ret;
> +		}
>  	}
>  
>  	return ret;
> -- 
> 2.31.1
>