Re: [PATCH v2 3/3] mmc: mediatek: add support for SDIO eint irq

From: Andy Shevchenko
Date: Fri Jan 14 2022 - 08:40:34 EST


On Fri, Jan 14, 2022 at 11:06:36AM +0800, Axe Yang wrote:
> On Thu, 2022-01-13 at 12:47 +0200, Andy Shevchenko wrote:
> > On Thu, Jan 13, 2022 at 03:58:52PM +0800, Axe Yang wrote:
> > > On Tue, 2022-01-11 at 13:46 +0200, Andy Shevchenko wrote:
> > > > On Tue, Jan 11, 2022 at 09:40:46AM +0800, Axe Yang wrote:

...

> > > > > + host->pins_eint = pinctrl_lookup_state(host-
> > > > > >pinctrl,
> > > > > "state_eint");
> > > > > + if (IS_ERR(host->pins_eint)) {
> > > > > + dev_dbg(&pdev->dev, "Cannot find
> > > > > pinctrl
> > > > > eint!\n");
> > > >
> > > > In debug mode of pin control this will bring a duplicate message.
> > >
> > > Can you explain more about this comment?
> > > I don't understand what the 'duplicate message' refers for.
> >
> > Have you chance to read the implementation of pinctrl_lookup_state()?
>
> I have read pinctrl_lookup_state(), and if the 'duplicate message' you
> were talking about is 'using pinctrl dummy state...':
>
> https://elixir.bootlin.com/linux/latest/source/drivers/pinctrl/core.c#L1214
> No, this message will not appear in debug mode if pins_eint not found
> because pinctrl_dummy_state is always FALSE. MTK Soc do not need dummy
> state support.

I see, thanks for explanation. I'm wondering how good or bad would be
the idea of adding a debug message on the !pinctrl_dummy_state case.

--
With Best Regards,
Andy Shevchenko