Re: [PATCH 0/3] gpio: Fix VLA removal fallout

[Linus Walleij <linus.walleij@xxxxxxxxxx>]

Hi Geert,

Thanks for the patches!

On Wed, Sep 5, 2018 at 11:23 AM Geert Uytterhoeven
<geert+renesas@xxxxxxxxx> wrote:

> This patch series fixes various (mostly harmless) issues introduced by
> commit 3027743f83f867d8 ("gpio: Remove VLA from gpiolib").
>
> As per the "one patch should fix one issue"-policy, this series contains 3
> patches, although they all have the same Fixes: tag.
>
> W.r.t. propagating errors: while gpiod_set_array_value_complex() and its
> callers can now return an error code, this is currently limited to -ENOMEM.
> Actual failures setting a GPIO output value cannot be propagated, as
> gpio_chip.set() still returns void.  Do you want to change that?
> E.g. gen_74x164_set_value() can fail.
>
> Feel free to fold patches if deemed appropriate.

What I want to know is if these patches drive a truck through Janusz patch
set augmenting the array functions that I definately also plan to merge for
this kernel cycle.

Issues should be fixed of course, but if some of them already disappear
if I apply Janusz patches, I'd rather postpone ... is it going to be hard
to redo the cleanups on top of his patches?

Yours,
Linus Walleij