Message ID | 20220409164556.2832782-1-bryan.odonoghue@linaro.org |
---|---|
Headers | show |
Series | Add camss to SM8250 dtsi | expand |
On 12/04/2022 03:04, Bjorn Andersson wrote: > On Sat 09 Apr 11:45 CDT 2022, Bryan O'Donoghue wrote: > >> There are a number of pin mux descriptions for the CAMSS block which are >> defined downstream. >> >> Bring in the definitions now. Subsequent changes will optionally activate >> those muxes for different sm8250 based boards. >> > > I'm not able to spot where these are referenced, are they not used or am > I just failing the search today? Used here: https://git.linaro.org/people/bryan.odonoghue/kernel.git/commit/?h=br-v5.19b%2brb5-dts%2bsensors&id=96d26b916efca18450ff67d385d846b50d8eb7e1 I'll send out the RB5 patch once I get regulator_bulk_*() added to the imx412 driver https://git.linaro.org/people/bryan.odonoghue/kernel.git/commit/?h=br-v5.19b%2brb5-dts%2bsensors&id=e11966d8101350d0e76eb3e1520165864bdbc296 > > I would prefer that we don't sprinkle pinctrl states that potentially > might be used at a later time... > >> Signed-off-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> >> --- >> arch/arm64/boot/dts/qcom/sm8250.dtsi | 471 +++++++++++++++++++++++++++ >> 1 file changed, 471 insertions(+) >> >> diff --git a/arch/arm64/boot/dts/qcom/sm8250.dtsi b/arch/arm64/boot/dts/qcom/sm8250.dtsi >> index 3d6c617de045..91ed079edbf7 100644 >> --- a/arch/arm64/boot/dts/qcom/sm8250.dtsi >> +++ b/arch/arm64/boot/dts/qcom/sm8250.dtsi >> @@ -3687,6 +3687,477 @@ tlmm: pinctrl@f100000 { >> gpio-ranges = <&tlmm 0 0 181>; >> wakeup-parent = <&pdc>; >> >> + cam_sensor_mclk0_active: cam_sensor_mclk0_active { > > Please don't use '_' in the node name. > >> + /* MCLK0 */ > > I would prefer that we don't inherit the mux/config split upstream. > > You can either do: > > cam-sensor-mclk0-active-state { > pins = ; > function = ; > bias... > }; > > or: > > cam-sensor-mclk0-active-state { > mclk0 { > pins = ; > function = ; > bias... > }; > }; > > And depending on how this state is actually used you could lump together > the configuration of several pins - with different properties - in the > same state node. Sure np --- bod