Re: [PATCH v2 0/3] pinctrl: fix possible memory leak when pinctrl_enable() fails

[Dan Carpenter <dan.carpenter@xxxxxxxxxx>]

On Thu, Jun 06, 2024 at 10:37:01AM +0800, Yang Yingliang wrote:
> In devm_pinctrl_register(), if pinctrl_enable() fails in pinctrl_register(),
> the "pctldev" has not been added to dev resources, so devm_pinctrl_dev_release()
> can not be called, it leads memory leak.
> 
> And some driver calls pinctrl_register_and_init() which is not devm_ managed,
> it also leads memory leak if pinctrl_enable() fails.
> 
> To fix this, introduce pinctrl_uninit_controller(), call it in the error path to
> free memory and replace pinctrl_register_and_init with devm_pinctrl_register_and_init.
> 
> v1 -> v2:
>   Check pinctrl_enable() return value in pcs_probe() in patch #2
> 
> Yang Yingliang (3):
>   pinctrl: core: fix possible memory leak when pinctrl_enable() fails
>   pinctrl: single: fix possible memory leak when pinctrl_enable() fails
>   pinctrl: ti: ti-iodelay: fix possible memory leak when
>     pinctrl_enable() fails

Thanks so much!

Reviewed-by: Dan Carpenter <dan.carpenter@xxxxxxxxxx>

regards,
dan carpenter