Re: [PATCH v3 09/11] mmc: add PCI dependency into IOSF_MBI

[Sinan Kaya]

On Wed, Dec 26, 2018 at 12:41 PM Rafael J. Wysocki <rafael@xxxxxxxxxx> wrote:
>
> On Mon, Dec 24, 2018 at 12:26 AM Sinan Kaya <okaya@xxxxxxxxxx> wrote:
> >
> > Select IOSF_MBI only when PCI is set.
> >
> > Signed-off-by: Sinan Kaya <okaya@xxxxxxxxxx>
> > ---
> >  drivers/mmc/host/Kconfig | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/drivers/mmc/host/Kconfig b/drivers/mmc/host/Kconfig
> > index e26b8145efb3..4c5f037b246d 100644
> > --- a/drivers/mmc/host/Kconfig
> > +++ b/drivers/mmc/host/Kconfig
> > @@ -117,7 +117,7 @@ config MMC_RICOH_MMC
> >  config MMC_SDHCI_ACPI
> >         tristate "SDHCI support for ACPI enumerated SDHCI controllers"
> >         depends on MMC_SDHCI && ACPI
>
> It should depend on PCI as well, shouldn't it?
>

I don't think so. The PCI dependency is at IOSF_MBI code so code
should select IOSF_MBI only when x86 and pci is defined as below.

This driver does not depend on PCI.

> As it stands with the patch applied, randconfig can still select it if
> PCI is unset AFAICS, can't it?

True, we also should be careful about separating ACPI from PCI.
Someone should be able to define a sdhci acpi object without PCI.

What do you think about this?

>
> > -       select IOSF_MBI if X86
> > +       select IOSF_MBI if (X86 && PCI)
> >         help
> >           This selects support for ACPI enumerated SDHCI controllers,
> >           identified by ACPI Compatibility ID PNP0D40 or specific
> > --