Re: [PATCH v3 07/11] gpiolib: replace find_chip_by_name() with gpio_device_find_by_label()

[Bartosz Golaszewski <brgl@xxxxxxxx>]

On Mon, Sep 18, 2023 at 9:23 AM Andy Shevchenko
<andriy.shevchenko@xxxxxxxxxxxxxxx> wrote:
>
> On Fri, Sep 15, 2023 at 05:03:22PM +0200, Bartosz Golaszewski wrote:
> > From: Bartosz Golaszewski <bartosz.golaszewski@xxxxxxxxxx>
> >
> > Remove all remaining uses of find_chip_by_name() (and subsequently:
> > gpiochip_find()) from gpiolib.c and use the new
> > gpio_device_find_by_label() instead.
>
> ...
>
> > -static int gpiochip_match_name(struct gpio_chip *gc, void *data)
> > -{
> > -     const char *name = data;
> > -
> > -     return !strcmp(gc->label, name);
>
> And this we had no check for the label being NULL...
>

Yeah, it was wrong. But maybe all kernel users already do assign it in
which case we should be safe just adding a check in
gpiochip_add_data_with_key() that would return EINVAL if they don't.

> ...
>
> >       for (p = &table->table[0]; p->key; p++) {
> > +             struct gpio_device *gdev __free(gpio_device_put) = NULL;
>
> > +             gdev = gpio_device_find_by_label(p->key);
> > +             if (!gdev) {
>
> I haven't got the fix for gpio-sim, shouldn't we have the same here, i.e.
> definition being done together with the assignment when __free() is in use?
>

It should but I only got yelled at by Linus under the gpio-sim patch
after I sent this one.

Bart

> >               }
>
> --
> With Best Regards,
> Andy Shevchenko
>
>