Re: [PATCH v7 0/6] gpio: Add GPIO Aggregator

[Geert Uytterhoeven]

Hi Andy,

On Wed, May 20, 2020 at 2:14 PM Andy Shevchenko
<address@hidden> wrote:
> On Mon, May 11, 2020 at 04:52:51PM +0200, Geert Uytterhoeven wrote:
> > GPIO controllers are exported to userspace using /dev/gpiochip*
> > character devices.  Access control to these devices is provided by
> > standard UNIX file system permissions, on an all-or-nothing basis:
> > either a GPIO controller is accessible for a user, or it is not.
> > Currently no mechanism exists to control access to individual GPIOs.
> >
> > Hence this adds a GPIO driver to aggregate existing GPIOs, and expose
> > them as a new gpiochip.  This is useful for implementing access control,
> > and assigning a set of GPIOs to a specific user.  Furthermore, this
> > simplifies and hardens exporting GPIOs to a virtual machine, as the VM
> > can just grab the full GPIO controller, and no longer needs to care
> > about which GPIOs to grab and which not, reducing the attack surface.
> > This has been implemented for ARM virt in QEMU[1].
> >
> > Recently, other use cases have been discovered[2], like describing
> > simple GPIO-operated devices in DT, and using the GPIO Aggregator as a
> > generic GPIO driver for userspace, which is useful for industrial
> > control.
> >
> > Note that the first patch of this series ("i2c: i801: Use GPIO_LOOKUP()
> > helper macro") has been applied to i2c/for-next.
>
> Sorry for late reply, recently noticed this nice idea.
> The comment I have is, please, can we reuse bitmap parse algorithm and syntax?
> We have too many different formats and parsers in the kernel and bitmap's one
> seems suitable here.

Thank you, I wasn't aware of that.

Which one do you mean? The documentation seems to be confusing,
and incomplete.
My first guess was bitmap_parse(), but that one assumes hex values?
And given it processes the unsigned long bitmap in u32 chunks, I guess
it doesn't work as expected on big-endian 64-bit?

bitmap_parselist() looks more suitable, and the format seems to be
compatible with what's currently used, so it won't change ABI.
Is that the one you propose?

> (Despite other small clean ups, like strstrip() use)

Aka strim()? There are too many of them, to know all of them by heart ;-)

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- address@hidden

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