Re: [PATCH 2/2] gpio: aggregator: Use bitmap_parselist() for parsing GPIO offsets

[Geert Uytterhoeven]

Hi Andy,

On Wed, Jun 24, 2020 at 3:41 PM Andy Shevchenko
<andy.shevchenko@xxxxxxxxx> wrote:
> On Wed, Jun 24, 2020 at 3:16 PM Bartosz Golaszewski
> <bgolaszewski@xxxxxxxxxxxx> wrote:
> > wt., 23 cze 2020 o 16:57 Geert Uytterhoeven <geert+renesas@xxxxxxxxx>
> > napisaÅ(a):
> > >
> > > Replace the custom code to parse GPIO offsets and/or GPIO offset ranges
> > > by a call to bitmap_parselist(), and an iteration over the returned bit
> > > mask.
> > >
> > > This should have no impact on the format of the configuration parameters
> > > written to the "new_device" virtual file in sysfs.
> > >
> > > Suggested-by: Andy Shevchenko <andy.shevchenko@xxxxxxxxx>
> > > Signed-off-by: Geert Uytterhoeven <geert+renesas@xxxxxxxxx>
> > > ---
> > > I'm not super happy with the mask[] array, which is on the stack.
> > > But there is no real limit on the number of GPIO lines provided by a
> > > single gpiochip, except for the global ARCH_NR_GPIOS.
> >
> > Why not allocate it with bitmap_zalloc() then?

Bartosz: Thanks, good idea!

> I haven't got the original messages yet, so my thought is to actually
> extract a helper from
> gpiod_get_array_value_complex() or gpiod_set_array_value_complex() for
> bitmap allocation.
> But I didn't check if it's suitable here. So, bitmap_zalloc() would be helpful.

/me confused

This function is not about getting/setting multiple GPIO lines in a gpiochip,
but about parsing a list of numbers and ranges. No gpiochip is involved.

original message:
https://lore.kernel.org/r/20200623145748.28877-3-geert+renesas@xxxxxxxxx

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds