Re: [PATCH] gpio: Allow user to customise maximum number of GPIOs

From: Davide Ciminaghi
Date: Wed Aug 31 2022 - 17:49:14 EST


On Thu, Sep 01, 2022 at 12:07:26AM +0300, Andy Shevchenko wrote:
> On Wed, Aug 31, 2022 at 5:14 PM Davide Ciminaghi <ciminaghi@xxxxxxxxx> wrote:
> > On Wed, Aug 31, 2022 at 03:32:25PM +0200, Linus Walleij wrote:
> > > On Tue, Aug 30, 2022 at 9:58 AM Davide Ciminaghi <ciminaghi@xxxxxxxxx> wrote:
> > >
> > > > the sta2x11 was a chip containing AMBA peripherals and a PCIe to AMBA bridge
> > > > (it is still in production as far as I know, but deprecated for new designs).
> > > > It would typically be installed on x86 machines, so you needed to build and
> > > > run AMBA drivers in an x86 environment. The original drivers we started from
> > > > had platform data, but then we were told to switch to DTS.
> > >
> > > For the record I think that was bad advice, I hope it wasn't me.
> > > But the world was different back then I suppose.
> > > Adding DTS to x86 which is inherently ACPI is not a good idea.
> > > Especially if you look at how SBSA ACPI UARTS were done
> > > in drivers/tty/serial/amba-pl011.c.
> > >
> > now that I think of it, ACPI was also listed as a possible choice, but the
> > problem was that we didn't know much about ACPI, and took the DTS way.
> > So there was no bad advice, just fear of the unknown :-)
>
> Feel free to ask, we have experts in the mailing list(s).
>
Thanks ! I'll keep that in mind in case I need some ACPI advice.
I'm afraid it's too late for the sta2x11, though. Its kernel is now a
downstream one, and it's been freezed, as the SOC has been deprecated
for new designs.
As Alessandro said, we'll submit (or ack) patches for removal.



Thanks again and regards
Davide