RE: [PATCH v4] arm64: dts: qcom: sc7180: Add WCN3990 WLAN module device node

[pillair]

Hi Matthias,

Please find my replies inline.


> -----Original Message-----
> From: Matthias Kaehlcke <mka@xxxxxxxxxxxx>
> Sent: Wednesday, January 29, 2020 12:50 AM
> To: Rakesh Pillai <pillair@xxxxxxxxxxxxxx>
> Cc: devicetree@xxxxxxxxxxxxxxx; linux-arm-kernel@xxxxxxxxxxxxxxxxxxx; linux-
> kernel@xxxxxxxxxxxxxxx; linux-arm-msm@xxxxxxxxxxxxxxx
> Subject: Re: [PATCH v4] arm64: dts: qcom: sc7180: Add WCN3990 WLAN
> module device node
> 
> Hi Rakesh,
> 
> On Tue, Jan 28, 2020 at 04:03:37PM +0530, Rakesh Pillai wrote:
> > Add device node for the ath10k SNOC platform driver probe
> > and add resources required for WCN3990 on sc7180 soc.
> >
> > Signed-off-by: Rakesh Pillai <pillair@xxxxxxxxxxxxxx>
> > ---
> >  arch/arm64/boot/dts/qcom/sc7180-idp.dts |  5 +++++
> >  arch/arm64/boot/dts/qcom/sc7180.dtsi    | 28
> ++++++++++++++++++++++++++++
> >  2 files changed, 33 insertions(+)
> >
> > diff --git a/arch/arm64/boot/dts/qcom/sc7180-idp.dts
> b/arch/arm64/boot/dts/qcom/sc7180-idp.dts
> > index 189254f..151b489 100644
> > --- a/arch/arm64/boot/dts/qcom/sc7180-idp.dts
> > +++ b/arch/arm64/boot/dts/qcom/sc7180-idp.dts
> > @@ -248,6 +248,11 @@
> >  	status = "okay";
> >  };
> >
> > +&wifi {
> > +	status = "okay";
> > +	qcom,msa-fixed-perm;
> > +};
> > +
> >  /* PINCTRL - additions to nodes defined in sc7180.dtsi */
> >
> >  &qup_i2c2_default {
> > diff --git a/arch/arm64/boot/dts/qcom/sc7180.dtsi
> b/arch/arm64/boot/dts/qcom/sc7180.dtsi
> > index 666e9b9..7efb97f 100644
> > --- a/arch/arm64/boot/dts/qcom/sc7180.dtsi
> > +++ b/arch/arm64/boot/dts/qcom/sc7180.dtsi
> > @@ -42,6 +42,12 @@
> >  			compatible = "qcom,cmd-db";
> >  			no-map;
> >  		};
> > +
> > +		wlan_fw_mem: memory@93900000 {
> > +			compatible = "removed-dma-pool";
> > +			no-map;
> > +			reg = <0 0x93900000 0 0x200000>;
> > +		};
> >  	};
> 
> This part doesn't apply cleanly on qcom/for-next, looks like you have to
> rebase.

[Rakesh]: let me rebase it and send out the next version.

> 
> >  	cpus {
> > @@ -1119,6 +1125,28 @@
> >  				#clock-cells = <1>;
> >  			};
> >  		};
> > +
> > +		wifi: wifi@18800000 {
> 
> You added this node at the end of the file, outside of the 'soc' node.
> It should be inside the 'soc' node, the sub-nodes are ordered by address,
> so (currently) this node should be inserted after 'cpufreq@18323000'.

[Rakesh] Will update this in the next patchset. (Possibly this was caused due to rebase conflicts)


> 
> > +			compatible = "qcom,wcn3990-wifi";
> > +			reg = <0 0x18800000 0 0x800000>;
> > +			reg-names = "membase";
> > +			iommus = <&apps_smmu 0xC0 0x1>;
> 
> nit: the convention is to use lowercase characters for hex adresses.

[Rakesh]: Will change it.

> 
> > +			interrupts =
> > +				<GIC_SPI 414 IRQ_TYPE_LEVEL_HIGH /* CE0
> */ >,
> > +				<GIC_SPI 415 IRQ_TYPE_LEVEL_HIGH /* CE1
> */ >,
> > +				<GIC_SPI 416 IRQ_TYPE_LEVEL_HIGH /* CE2
> */ >,
> > +				<GIC_SPI 417 IRQ_TYPE_LEVEL_HIGH /* CE3
> */ >,
> > +				<GIC_SPI 418 IRQ_TYPE_LEVEL_HIGH /* CE4
> */ >,
> > +				<GIC_SPI 419 IRQ_TYPE_LEVEL_HIGH /* CE5
> */ >,
> > +				<GIC_SPI 420 IRQ_TYPE_LEVEL_HIGH /* CE6
> */ >,
> > +				<GIC_SPI 421 IRQ_TYPE_LEVEL_HIGH /* CE7
> */ >,
> > +				<GIC_SPI 422 IRQ_TYPE_LEVEL_HIGH /* CE8
> */ >,
> > +				<GIC_SPI 423 IRQ_TYPE_LEVEL_HIGH /* CE9
> */ >,
> > +				<GIC_SPI 424 IRQ_TYPE_LEVEL_HIGH /* CE10
> */>,
> > +				<GIC_SPI 425 IRQ_TYPE_LEVEL_HIGH /* CE11
> */>;
> 
> not sure these 'CEx' comments after each interrupt add much value. What
> does
> 'CE' stand for in the first place?

[Rakesh]:   CE stands for Copy engine. These are the copy engine interrupts, hence the comments.

> 
> Thanks
> 
> Matthias