On 08/20, Neil Armstrong wrote:
> On 08/18/2016 08:08 PM, Stephen Boyd wrote:
> > On 08/18, Neil Armstrong wrote:
> > 
> > 
> >> +
> >> +          gsbi2: gsbi@16100000 {
> >> +                  compatible = "qcom,gsbi-v1.0.0";
> >> +                  cell-index = <2>;
> >> +                  reg = <0x16100000 0x100>;
> >> +                  clocks = <&gcc GSBI2_H_CLK>;
> >> +                  clock-names = "iface";
> >> +                  status = "disabled";
> >> +                  #address-cells = <1>;
> >> +                  #size-cells = <1>;
> >> +                  ranges;
> >> +
> >> +                  gsbi2_i2c: i2c@16180000 {
> >> +                          compatible = "qcom,i2c-qup-v1.1.1";
> >> +                          #address-cells = <1>;
> >> +                          #size-cells = <0>;
> >> +                          reg = <0x16180000 0x1000>;
> >> +                          interrupts = <GIC_SPI 149 IRQ_TYPE_NONE>;
> > 
> > There should be a trigger type... high perhaps?
> 
> Well, I do not know, and other qcom dtsi has 0x0 or NONE here...

We should fix those. The driver uses IRQF_TRIGGER_HIGH so it
seems like we should put that here as well.

> 
> > 
> >> +
> >> +                          clocks = <&gcc GSBI2_QUP_CLK>, <&gcc 
> >> GSBI2_H_CLK>;
> >> +                          clock-names = "core", "iface";
> >> +                          status = "disabled";
> >> +                  };
> >> +          };
> >> +
> >> +
> >> +          qcom,ssbi@500000 {
> >> +                  compatible = "qcom,ssbi";
> >> +                  reg = <0x500000 0x1000>;
> >> +                  qcom,controller-type = "pmic-arbiter";
> >> +
> >> +                  pmicintc: pmic@0 {
> >> +                          compatible = "qcom,pm8018", "qcom,pm8921";
> > 
> > I know that DT specifies most specific compatible first, but when
> > the generic compatible is part number specific as well it never
> > feels right to me. I guess I'll have to get over this.
> > 
> >> +                          interrupts = <GIC_PPI 226 IRQ_TYPE_NONE>;
> >> +                          #interrupt-cells = <2>;
> >> +                          interrupt-controller;
> >> +                          #address-cells = <1>;
> >> +                          #size-cells = <0>;
> > 
> > Can we have interrupt-parent = <&pmicintc> here instead of in
> > every node?
> 
> No since the pmic node interrupt parent is the GIC.

Ah ok. Silly me.

> > 
> >> +                  compatible = "qcom,bam-v1.3.0";
> >> +                  reg = <0x12142000 0x8000>;
> >> +                  interrupts = <GIC_SPI 97 IRQ_TYPE_NONE>;
> > 
> > There should really be some flags.
> 
> Same as for the qup and uartdm, other qcom dtsi left 0x0 or NONE here...

The driver uses IRQF_TRIGGER_HIGH here, so use that in dtsi?

-- 
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project

Reply via email to