On 5/28/24 14:45, Marc Gonzalez wrote:
On 28/05/2024 14:31, Konrad Dybcio wrote:
[...]
+ hdmi_cec_default: hdmi-cec-default-state {
+ pins = "gpio31";
+ function = "hdmi_cec";
+ drive-strength = <2>;
+ bias-pull-up;
+ };
+
+ hdmi_cec_sleep: hdmi-cec-sleep-state {
+ pins = "gpio31";
+ function = "hdmi_cec";
+ drive-strength = <2>;
+ bias-pull-up;
+ };
It's super strange that both states are identical. Usually, the pull-up
is disabled and the GPIO is unmuxed (i.e. function = "gpio"). If that's
not the case for whatever reason, you can drop the sleep variants and
simply reference the leftover one from both pinctrl-0 and pinctrl-1.
Patch is a direct translation of the vendor code:
https://git.codelinaro.org/clo/la/kernel/msm-4.4/-/blob/caf_migration/kernel.lnx.4.4.r38-rel/arch/arm/boot/dts/qcom/msm8998-pinctrl.dtsi#L2324-2400
I suppose it wouldn't be the first time that vendor code
is doing something odd.
Though, I'm a bit confused why you would single out hdmi-cec
when hdmi_ddc is the same?
As in, me in the above message or the vendor devicetree?
If the former, it's just an example
If the latter, the muxing function differs so they must have their
own separate nodes
Konrad
