Re: [PATCH v2 7/7] arm64: dts: qcom: lemans: Enable lemans-sa8255p-ride board support

[Shawn Guo]

On Tue, Oct 20, 2026 at 02:16:27PM +0200, Konrad Dybcio wrote:
> On 10/15/26 4:37 AM, Shawn Guo wrote:
> > From: Nikunj Kela <quic_nkela@xxxxxxxxxxx>
> > 
> > Add initial device tree support for the Lemans SA8255P Ride board. Enable
> > UFS storage, UART10 as the serial console, PCIe EP interfaces, thermal
> > zones for PMM8654AU sensors, and SCMI protocol nodes.
> > 
> > Co-developed-by: Shazad Hussain <shazad.hussain@xxxxxxxxxxxxxxxx>
> > Signed-off-by: Shazad Hussain <shazad.hussain@xxxxxxxxxxxxxxxx>
> > Signed-off-by: Nikunj Kela <quic_nkela@xxxxxxxxxxx>
> > Co-developed-by: Deepti Jaggi <deepti.jaggi@xxxxxxxxxxxxxxxx>
> > Signed-off-by: Deepti Jaggi <deepti.jaggi@xxxxxxxxxxxxxxxx>
> > Reviewed-by: Bartosz Golaszewski <bartosz.golaszewski@xxxxxxxxxxxxxxxx>
> > Signed-off-by: Shawn Guo <shengchao.guo@xxxxxxxxxxxxxxxx>
> > ---
> 
> [...]
> 
> > +	thermal-zones {
> > +		pmm8654au_0_thermal: pm8255-0-thermal {
> > +			polling-delay-passive = <100>;
> 
> Do we need the polling, is there no interrupt?

Yes, we need polling. The sensors are exposed via the SCMI sensor
protocol, and the SCMI server on SA8255P doesn't seem to support trip
points. Also on the kernel side, scmi-hwmon driver implements only
.get_temp for the thermal zone and doesn't handle trip-point events
either. So polling is the only way to evaluate the trips.

> 
> [...]
> 
> 
> > +&pcie0_ep {
> > +	pinctrl-names = "default";
> > +	pinctrl-0 = <&pcie0_ep_clkreq_default &pcie0_ep_perst_default
> > +		    &pcie0_ep_wake_default>;
> > +	reset-gpios = <&tlmm 2 GPIO_ACTIVE_LOW>;
> > +	wake-gpios = <&tlmm 0 GPIO_ACTIVE_HIGH>;
> 
> This GPIO reference feels like a layering violation - the previous
> commit promised TLMM is abstracted away

I will reword commit message of patch 5 to clarify that kernel-controlled
TLMM GPIOs are not abstracted by firmware.

Shawn