Re: Should irq_enable/disable be flagging/unflagging gpio lines used as irq?

[Linus Walleij <linus.walleij@linaro.org>]

On Thu, Jul 13, 2017 at 10:17 PM, Davide Hug <d@videhug.ch> wrote:

> Thanks. I probably should have pointed out that I'm just getting started.
> I thought this driver that does not load would be a good place to do that. But
> obviously a "completely broken" driver that "breaks with the current semantics
> of the GPIO interrupt" is the wrong place to get started and learn how things
> should be done.
>
> So this driver should probably be rewritten using timer based polling and the
> descriptor based gpio interface. But since it is a really specific driver for
> sensors that are already discontinued it probably doesn't make much sense. And
> I'm not yet up to the task anyway.

I'm not so sure about that. You might be the only one with the right hardware
to fix the problem. Also it is never our intention to discourage anyone to
hack on the kernel.

If you have this hardware and know what you want to get out of it, go for it.

> Are the drivers listed in Documentation/gpio/drivers-on-gpio.txt good examples
> of GPIO consumer drivers?

I think some, I am fixing others, anything just including
<linux/gpio/consumer.h>
and no other GPIO header are usually good examples.

Yours,
Linus Walleij