Re: [PATCH v2 2/3] arm64: dts: qcom: sc7280: add lpass lpi pin controller node

From: Srinivasa Rao Mandadapu
Date: Thu Feb 10 2022 - 09:28:05 EST



On 2/10/2022 5:35 AM, Stephen Boyd wrote:
Thanks for Your time Stephen!!!
Quoting Srinivasa Rao Mandadapu (2022-02-09 06:01:21)
On 2/9/2022 2:41 AM, Stephen Boyd wrote:
Quoting Srinivasa Rao Mandadapu (2022-02-08 07:34:13)
+ data {
+ pins = "gpio7";
+ function = "dmic1_data";
+ drive-strength = <8>;
+ input-enable;
+ };
+ };
+
+ dmic01_sleep: dmic01-sleep-pins {
+ clk {
+ pins = "gpio6";
+ function = "dmic1_clk";
+ drive-strength = <2>;
+ bias-disable;
+ output-low;
+ };
+
+ data {
+ pins = "gpio7";
+ function = "dmic1_data";
+ drive-strength = <2>;
+ pull-down;
+ input-enable;
Why does input-enable matter? It's not a gpio.
Actually the same is fallowed in sm8250.dtsi. Verified without it and
working fine. Need take call on it.
Is that because the pin is already an input by default? What does gpio
debugfs say for this pin? Does it also work if you make it
output-low/output-high here? I thought that the gpio itself isn't muxed
out to the pad unless the function is "gpio" so I hope the input/output
settings have no effect here.

Pin is in by default. debugfs says

gpio7 : in 1 8mA no pull

verified in downstream code also. Same is followed there also.