Re: [PATCH net-next v8 6/9] net: txgbe: Support GPIO to SFP socket

[Andy Shevchenko <andy.shevchenko@xxxxxxxxx>]

I just realized that we are discussing all this without GPIO
maintainers to be involved!
Cc'ed to Linus and Bart for their valuable opinions / comments.

(TL;DR: GPIO regmap seems need some fixes)

On Tue, May 23, 2023 at 2:07 PM Andy Shevchenko
<andy.shevchenko@xxxxxxxxx> wrote:
> On Tue, May 23, 2023 at 12:57 PM Jiawen Wu <jiawenwu@xxxxxxxxxxxxxx> wrote:
> > > > Anyway it is a bit complicated, could I use this version of GPIO implementation if
> > > > it's really tough?
> > >
> > > It's possible but from GPIO subsystem point of view it's discouraged
> > > as long as there is no technical impediment to go the regmap way.
> >
> > After these days of trying, I guess there are still some bugs on gpio - regmap - irq.
> > It looks like an compatibility issue with gpio_irq_chip and regmap_irq_chip (My rough
> > fixes seems to work).
> >
> > Other than that, it seems to be no way to add interrupt trigger in regmap_irq_thread(),
> > to solve the both-edge problem for my hardware.
> >
> > I'd be willing to use gpio-regmap if above issues worked out, I learned IRQ controller,
> > IRQ domain, etc. , after all.
>
> And thank you for all this!
> Now you may suggest the fixes to the GPIO regmap with all your
> knowledge of the area.
>
> > But if not, I'd like to implement GPIO in the original way,
> > it was tested to work. May I? Thanks for all your suggestions.
>
>
> --
> With Best Regards,
> Andy Shevchenko



--
With Best Regards,
Andy Shevchenko