Re: [PATCH v1] arm64: dts: qcom: qcs615: Add QUPv3 configuration

[Viken Dadhaniya]

On 10/26/2024 1:10 AM, Konrad Dybcio wrote:
> On 11.10.2024 12:33 PM, Viken Dadhaniya wrote:
> > Add DT support for QUPv3 Serial Engines.
> >
> > Co-developed-by: Mukesh Kumar Savaliya <quic_msavaliy@xxxxxxxxxxx>
> > Signed-off-by: Mukesh Kumar Savaliya <quic_msavaliy@xxxxxxxxxxx>
> > Signed-off-by: Viken Dadhaniya <quic_vdadhani@xxxxxxxxxxx>
> > ---
>
> [...]
>
> > +	qup_opp_table: opp-table-qup {
> > +		compatible = "operating-points-v2";
>
> opp-shared;

Added in v2.

> > +
> > +		opp-75000000 {
> > +			opp-hz = /bits/ 64 <75000000>;
> > +			required-opps = <&rpmhpd_opp_low_svs>;
> > +		};
> > +
> > +		opp-100000000 {
> > +			opp-hz = /bits/ 64 <100000000>;
> > +			required-opps = <&rpmhpd_opp_svs>;
> > +		};
> > +
> > +		opp-128000000 {
> > +			opp-hz = /bits/ 64 <128000000>;
> > +			required-opps = <&rpmhpd_opp_nom>;
> > +		};
> > +	};
> > +
> >   	psci {
> >   		compatible = "arm,psci-1.0";
> >   		method = "smc";
> > @@ -392,6 +427,24 @@
> >   			#size-cells = <1>;
> >   		};
> >   
> > +		gpi_dma0: qcom,gpi-dma@800000  {
> > +			compatible = "qcom,sdm845-gpi-dma";
>
> You must define a new compatible for qcs615, sdm845 is used as a fallback
> (so that we don't have to add new driver entries). You will however need
> to submit a separate dt-bindings change.

We have added sdm845 in compatible due to below comment in driver file

File: 
https://github.com/torvalds/linux/blob/81983758430957d9a5cb3333fe324fd70cf63e7e/drivers/dma/qcom/gpi.c#L2284

  /*
   * Do not grow the list for compatible devices. Instead use
   * qcom,sdm845-gpi-dma (for ee_offset = 0x0) or qcom,sm6350-gpi-dma
   * (for ee_offset = 0x10000).
   */

Do we still require new compatible for qcs615 ?

> > +			reg = <0x0 0x800000 0x0 0x60000>;
> > +			#dma-cells = <3>;
> > +			interrupts = <GIC_SPI 244 IRQ_TYPE_LEVEL_HIGH>,
> > +				     <GIC_SPI 245 IRQ_TYPE_LEVEL_HIGH>,
> > +				     <GIC_SPI 246 IRQ_TYPE_LEVEL_HIGH>,
> > +				     <GIC_SPI 247 IRQ_TYPE_LEVEL_HIGH>,
> > +				     <GIC_SPI 248 IRQ_TYPE_LEVEL_HIGH>,
> > +				     <GIC_SPI 249 IRQ_TYPE_LEVEL_HIGH>,
> > +				     <GIC_SPI 250 IRQ_TYPE_LEVEL_HIGH>,
> > +				     <GIC_SPI 251 IRQ_TYPE_LEVEL_HIGH>;
> > +			dma-channels = <8>;
> > +			dma-channel-mask = <0xf>;
> > +			iommus = <&apps_smmu 0xd6 0x0>;
> > +			status = "disabled";
>
> Any reason?

By default, we are disabling all nodes. In the target-specific file, we 
will update them based on the required use case.

> > +		};
> > +
> >   		qupv3_id_0: geniqup@8c0000 {
> >   			compatible = "qcom,geni-se-qup";
> >   			reg = <0x0 0x8c0000 0x0 0x6000>;
> > @@ -400,6 +453,7 @@
> >   				 <&gcc GCC_QUPV3_WRAP_0_S_AHB_CLK>;
> >   			clock-names = "m-ahb",
> >   				      "s-ahb";
> > +			iommus = <&apps_smmu 0xc3 0x0>;
>
> This looks like a separate fix

One QUP and UART (console) related change has been added as part of the 
base DTSI. Now, we are adding all remaining QUP and I2C/SPI/UART 
configurations with IOMMUs and other properties.

Is there any specific reason to make this a separate fix?

> >   			#address-cells = <2>;
> >   			#size-cells = <2>;
> >   			status = "disabled";
> > @@ -412,13 +466,377 @@
> >   				pinctrl-0 = <&qup_uart0_tx>, <&qup_uart0_rx>;
> >   				pinctrl-names = "default";
> >   				interrupts = <GIC_SPI 601 IRQ_TYPE_LEVEL_HIGH>;
> > -				interconnects = <&aggre1_noc MASTER_QUP_0 0
> > -						 &mc_virt SLAVE_EBI1 0>,
> > -						<&gem_noc MASTER_APPSS_PROC 0
> > -						 &config_noc SLAVE_QUP_0 0>;
> > +				interconnects = <&aggre1_noc MASTER_QUP_0 0 &mc_virt SLAVE_EBI1 0>,
> > +						<&gem_noc MASTER_APPSS_PROC 0 &config_noc SLAVE_QUP_0 0>;
>
> Why?
>
> Also, please use QCOM_ICC_TAG_ALWAYS instead of zeroes

Updated in v2.

> Konrad