Re: [net-next 3/3] arm64: dts: renesas: r8a779g0: Add gPTP node

["Niklas Söderlund" <niklas.soderlund+renesas@ragnatech.se>]

Hello,

On 2026-06-10 21:57:40 +0000, sashiko-bot@kernel.org wrote:
> Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
> - [High] Device Tree Resource Conflict due to overlapping memory, clock, and reset regions between `tsn0` and `gptp`.
> --
> 
> commit 225abba0fdcd1f7e4ebd535f168ba1447d871a7e
> Author: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
> 
> arm64: dts: renesas: r8a779g0: Add gPTP node
> 
> The gPTP module is shared between the RAVB and RTSN Ethernet devices on
> the SoC.
> 
> > diff --git a/arch/arm64/boot/dts/renesas/r8a779g0.dtsi b/arch/arm64/boot/dts/renesas/r8a779g0.dtsi
> > index 82a7278836e57..758c4ca0ae5f2 100644
> > --- a/arch/arm64/boot/dts/renesas/r8a779g0.dtsi
> > +++ b/arch/arm64/boot/dts/renesas/r8a779g0.dtsi
> > @@ -603,6 +603,15 @@ tsn0: ethernet@e6460000 {
> >  			status = "disabled";
> >  		};
> >  
> > +		gptp: gptp@e6449000 {
> > +			compatible = "renesas,r8a779g0-gptp", "renesas,rcar-gen4-gptp";
> > +			reg = <0 0xe6449000 0 0x500>;
> > +			clocks = <&cpg CPG_MOD 2723>;
> > +			power-domains = <&sysc R8A779G0_PD_ALWAYS_ON>;
> > +			resets = <&cpg 2723>;
> 
> [Severity: High]
> Will this introduce a regression by duplicating resources already claimed by
> the existing tsn0 node?

Yes.

That is the idea of this work. To break free the ptp memory region from 
the RTSN (on V4H) and RSWITCH (on S4) so it can be shared with RAVB.

This is discussed in the cover letter of the series.

> 
> Looking at the tsn0 node earlier in this file:
> 
> 		tsn0: ethernet@e6460000 {
> 			compatible = "renesas,r8a779g0-ethertsn", "renesas,rcar-gen4-ethertsn";
> 			reg = <0 0xe6460000 0 0x7000>,
> 			      <0 0xe6449000 0 0x500>;
> 			reg-names = "tsnes", "gptp";
> 			interrupts = <GIC_SPI 429 IRQ_TYPE_LEVEL_HIGH>,
> 				     <GIC_SPI 430 IRQ_TYPE_LEVEL_HIGH>;
> 			interrupt-names = "tx", "rx";
> 			clocks = <&cpg CPG_MOD 2723>;
> 			power-domains = <&sysc R8A779G0_PD_ALWAYS_ON>;
> 			resets = <&cpg 2723>;
> 			status = "disabled";
> 		};
> 
> It appears tsn0 already claims the 0xe6449000 memory region, CPG_MOD 2723 clock,
> and reset 2723.
> 
> If both drivers are enabled, will the second driver to probe fail when
> requesting these overlapping resources, such as receiving -EBUSY from
> devm_ioremap_resource() or request_mem_region()?
> 
> Should the overlapping resources be removed from the tsn0 node now that the
> gPTP module is represented as a standalone shared node?
> 
> > +			status = "disabled";
> > +		};
> > +
> >  		i2c0: i2c@e6500000 {
> 
> -- 
> Sashiko AI review · https://sashiko.dev/#/patchset/20260609215711.2960150-4-niklas.soderlund+renesas@ragnatech.se?part=1

-- 
Kind Regards,
Niklas Söderlund