Re: [PATCH v1 0/2] pinctrl: qcom: Make pinctrl drivers default y

[Sumit Garg <sumit.garg@kernel.org>]

On Fri, Dec 26, 2025 at 08:56:29AM -0600, Tom Rini wrote:
> On Fri, Dec 26, 2025 at 02:07:18PM +0530, Sumit Garg wrote:
> > On Mon, Dec 22, 2025 at 05:21:42PM +0530, Aswin Murugan wrote:
> > > This is a follow-up patch series to address review feedback on the
> > > QCS615 pinctrl driver addition:
> > > https://lore.kernel.org/u-boot/20251112164758.1560041-1-aswin.murugan@oss.qualcomm.com/
> > > 
> > > During review, it was suggested that since all Qualcomm pinctrl
> > > drivers are already within a menu conditional on ARCH_SNAPDRAGON ||
> > > ARCH_IPQ40XX, they should be made default y based on the appropriate
> > > architecture rather than requiring manual enablement in defconfig
> > > files.
> > > 
> > > This series implements that suggestion by:
> > > 1. Adding 'default y if ARCH_SNAPDRAGON' to all Snapdragon pinctrl
> > >    drivers and 'default y if ARCH_IPQ40XX' to all IPQ pinctrl
> > >    drivers in the Kconfig
> > 
> > I see the usefullness of this feature but in turn it will limit the
> > capability to reduce U-Boot size to an essential minimum. In scenarios
> > where there are contraints on flash size or on-chip RAM for U-Boot SPL.
> > I would rather favour something like following like:
> > 
> > 'default y if PINCTRL_QCOM_GENERIC'
> > 
> > where PINCTRL_QCOM_GENERIC can be enabled from generic Qcom defconfig.
> 
> If something is merely "default y" it can still be easily turned off.

Yeah true but then again in say any SPL defconfig every config option
needs to be turned off which is a similar thing you want to avoid now.

> If
> we want to make the "if ..." part more precise, OK.

Exactly, lets do it properly.

> But I don't want to
> go back to (nearly?) every snapdragon and ipq40xx defconfig being
> modified, as it was in an earlier iteration.

Sure, we don't want to go back but extend your suggestion with
PINCTRL_QCOM_GENERIC.

-Sumit