Re: [PATCH 1/2] gpio: omap: allow building the module with COMPILE_TEST=y

[Bartosz Golaszewski <brgl@xxxxxxxx>]

On Tue, Dec 3, 2024 at 10:54 PM Andrew Davis <afd@xxxxxx> wrote:
>
> On 12/3/24 2:36 PM, Bartosz Golaszewski wrote:
> > On Tue, Dec 3, 2024 at 7:41 PM Andrew Davis <afd@xxxxxx> wrote:
> >>
> >> On 12/3/24 10:41 AM, Bartosz Golaszewski wrote:
> >>> From: Bartosz Golaszewski <bartosz.golaszewski@xxxxxxxxxx>
> >>>
> >>> For better build coverage, allow building the gpio-omap driver with
> >>> COMPILE_TEST Kconfig option enabled.
> >>>
> >>> Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@xxxxxxxxxx>
> >>> ---
> >>>    drivers/gpio/Kconfig | 2 +-
> >>>    1 file changed, 1 insertion(+), 1 deletion(-)
> >>>
> >>> diff --git a/drivers/gpio/Kconfig b/drivers/gpio/Kconfig
> >>> index 56fee58e281e7..fb923ccd79028 100644
> >>> --- a/drivers/gpio/Kconfig
> >>> +++ b/drivers/gpio/Kconfig
> >>> @@ -530,7 +530,7 @@ config GPIO_OCTEON
> >>>    config GPIO_OMAP
> >>>        tristate "TI OMAP GPIO support" if ARCH_OMAP2PLUS || COMPILE_TEST
> >>>        default y if ARCH_OMAP
> >>> -     depends on ARM
> >>> +     depends on ARM || COMPILE_TEST
> >>
> >> Why do we have this depends on ARM at all? It already has that condition
> >> above on ARCH_OMAP2PLUS which limits to ARM outside of compile testing.
> >>
> >> And anything that selects ARCH_OMAP2PLUS also selects ARCH_OMAP, so we
> >> could just do this:
> >>
> >
> > I agree we can drop that bit.
> >
> >> --- a/drivers/gpio/Kconfig
> >> +++ b/drivers/gpio/Kconfig
> >> @@ -528,9 +528,9 @@ config GPIO_OCTEON
> >>             family of SOCs.
> >>
> >>    config GPIO_OMAP
> >> -       tristate "TI OMAP GPIO support" if ARCH_OMAP2PLUS || COMPILE_TEST
> >> -       default y if ARCH_OMAP
> >> -       depends on ARM
> >> +       tristate "TI OMAP GPIO support"
> >> +       default y
> >> +       depends on ARCH_OMAP2PLUS || COMPILE_TEST
> >
> > This would default to y with COMPILE_TEST. We definitely don't want
> > that. IMO it should be:
> >
> > tristate "TI OMAP GPIO support"
> > depends on ARCH_OMAP2PLUS || COMPILE_TEST
> > default y if ARCH_OMAP2PLUS
> >
>
> Looks good to me
>
> Andrew

Nah, this is incorrect, it doesn't build gpio-omap for
omap1_defconfig. I has to depend on ARCH_OMAP under which all omap
platforms fall and they all use this driver.

Bart