Re: [PATCH v4 1/4] nvmem: fix memory leak in error path

[Bartosz Golaszewski <bgolaszewski@xxxxxxxxxxxx>]

czw., 20 lut 2020 o 12:30 Srinivas Kandagatla
<srinivas.kandagatla@xxxxxxxxxx> napisał(a):
>
>
>
> On 20/02/2020 10:01, Bartosz Golaszewski wrote:
> > -     if (IS_ERR(nvmem->wp_gpio))
> > +     if (IS_ERR(nvmem->wp_gpio)) {
> > +             ida_simple_remove(&nvmem_ida, nvmem->id);
> > +             kfree(nvmem);
> >               return ERR_CAST(nvmem->wp_gpio);
> You freed nvmem just before this statement, how can nvmem->wp_gpio be
> still be valid?
>
> Are you able to test this changes at your end?
> Or
> these are just compile tested?
>

Sorry this was rushed, I will have access to the HW for testing tomorrow.

Bartosz