| Message ID | 20230712-squealer-walmart-9587342ddec1@wendy (mailing list archive) |
|---|---|
| State | Mainlined |
| Commit | fb22670b516aa2aaa8677d01ff85567ef7f14f1b |
| Delegated to: | Geert Uytterhoeven |
| Headers | show |
| Series | [v1] riscv: dts: renesas: clean up dtbs_check W=1 warning due to empty phy node | expand |
Hello! On 7/12/23 11:14 AM, Conor Dooley wrote: > dtbs_check w/ W=1 complains: > Warning (unit_address_vs_reg): /soc/ethernet@11c20000/ethernet-phy@7: node has a unit name, but no reg or ranges property > Warning (avoid_unnecessary_addr_size): /soc/ethernet@11c20000: unnecessary #address-cells/#size-cells without "ranges" or child "reg" property > > The ethernet@11c20000 node is guarded by an `#if (!SW_ET0_EN_N)` in > rzg2ul-smarc-som.dtsi, where the phy child node is added. In > rzfive-smarc-som.dtsi, the ethernet node is marked disabled & the > interrupt properties are deleted from the phy child node. As a result, > the produced dts looks like: > ethernet@11c20000 { > compatible = "renesas,r9a07g043-gbeth\0renesas,rzg2l-gbeth"; > /* snip */ > #address-cells = <0x01>; > #size-cells = <0x00>; > status = "disabled"; > > ethernet-phy@7 { > }; > }; > > Adding a corresponding `#if (!SW_ET0_EN_N)` around the node in > rzfive-smarc-som.dtsi avoids the complaint, as the empty child node is > not added: > ethernet@11c20000 { > compatible = "renesas,r9a07g043-gbeth\0renesas,rzg2l-gbeth"; > /* snip */ > #address-cells = <0x01>; > #size-cells = <0x00>; > status = "disabled"; > }; > > Signed-off-by: Conor Dooley <conor.dooley@microchip.com> [...] > diff --git a/arch/riscv/boot/dts/renesas/rzfive-smarc-som.dtsi b/arch/riscv/boot/dts/renesas/rzfive-smarc-som.dtsi > index d6f18754eb5d..c62debc7ca7e 100644 > --- a/arch/riscv/boot/dts/renesas/rzfive-smarc-som.dtsi > +++ b/arch/riscv/boot/dts/renesas/rzfive-smarc-som.dtsi > @@ -22,6 +22,7 @@ &dmac { > status = "disabled"; > }; > > +#if (!SW_ET0_EN_N) Are the parens really needed here? > ð0 { > status = "disabled"; > [...] MBR, Sergey
On Wed, Jul 12, 2023 at 10:15 AM Conor Dooley <conor.dooley@microchip.com> wrote: > dtbs_check w/ W=1 complains: > Warning (unit_address_vs_reg): /soc/ethernet@11c20000/ethernet-phy@7: node has a unit name, but no reg or ranges property > Warning (avoid_unnecessary_addr_size): /soc/ethernet@11c20000: unnecessary #address-cells/#size-cells without "ranges" or child "reg" property > > The ethernet@11c20000 node is guarded by an `#if (!SW_ET0_EN_N)` in > rzg2ul-smarc-som.dtsi, where the phy child node is added. In > rzfive-smarc-som.dtsi, the ethernet node is marked disabled & the > interrupt properties are deleted from the phy child node. As a result, > the produced dts looks like: > ethernet@11c20000 { > compatible = "renesas,r9a07g043-gbeth\0renesas,rzg2l-gbeth"; > /* snip */ > #address-cells = <0x01>; > #size-cells = <0x00>; > status = "disabled"; > > ethernet-phy@7 { > }; > }; > > Adding a corresponding `#if (!SW_ET0_EN_N)` around the node in > rzfive-smarc-som.dtsi avoids the complaint, as the empty child node is > not added: > ethernet@11c20000 { > compatible = "renesas,r9a07g043-gbeth\0renesas,rzg2l-gbeth"; > /* snip */ > #address-cells = <0x01>; > #size-cells = <0x00>; > status = "disabled"; > }; > > Signed-off-by: Conor Dooley <conor.dooley@microchip.com> Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be> i.e. will queue in renesas-devel for v6.6. Gr{oetje,eeting}s, Geert
Hi Sergei. On Wed, Jul 12, 2023 at 10:43 AM Sergei Shtylyov <sergei.shtylyov@gmail.com> wrote: > On 7/12/23 11:14 AM, Conor Dooley wrote: > > > dtbs_check w/ W=1 complains: > > Warning (unit_address_vs_reg): /soc/ethernet@11c20000/ethernet-phy@7: node has a unit name, but no reg or ranges property > > Warning (avoid_unnecessary_addr_size): /soc/ethernet@11c20000: unnecessary #address-cells/#size-cells without "ranges" or child "reg" property > > > > The ethernet@11c20000 node is guarded by an `#if (!SW_ET0_EN_N)` in > > rzg2ul-smarc-som.dtsi, where the phy child node is added. In > > rzfive-smarc-som.dtsi, the ethernet node is marked disabled & the > > interrupt properties are deleted from the phy child node. As a result, > > the produced dts looks like: > > ethernet@11c20000 { > > compatible = "renesas,r9a07g043-gbeth\0renesas,rzg2l-gbeth"; > > /* snip */ > > #address-cells = <0x01>; > > #size-cells = <0x00>; > > status = "disabled"; > > > > ethernet-phy@7 { > > }; > > }; > > > > Adding a corresponding `#if (!SW_ET0_EN_N)` around the node in > > rzfive-smarc-som.dtsi avoids the complaint, as the empty child node is > > not added: > > ethernet@11c20000 { > > compatible = "renesas,r9a07g043-gbeth\0renesas,rzg2l-gbeth"; > > /* snip */ > > #address-cells = <0x01>; > > #size-cells = <0x00>; > > status = "disabled"; > > }; > > > > Signed-off-by: Conor Dooley <conor.dooley@microchip.com> > [...] > > > diff --git a/arch/riscv/boot/dts/renesas/rzfive-smarc-som.dtsi b/arch/riscv/boot/dts/renesas/rzfive-smarc-som.dtsi > > index d6f18754eb5d..c62debc7ca7e 100644 > > --- a/arch/riscv/boot/dts/renesas/rzfive-smarc-som.dtsi > > +++ b/arch/riscv/boot/dts/renesas/rzfive-smarc-som.dtsi > > @@ -22,6 +22,7 @@ &dmac { > > status = "disabled"; > > }; > > > > +#if (!SW_ET0_EN_N) > > Are the parens really needed here? No they aren't. But this follows the existing style of the other users. > > > ð0 { > > status = "disabled"; > > > [...] Gr{oetje,eeting}s, Geert
diff --git a/arch/riscv/boot/dts/renesas/rzfive-smarc-som.dtsi b/arch/riscv/boot/dts/renesas/rzfive-smarc-som.dtsi index d6f18754eb5d..c62debc7ca7e 100644 --- a/arch/riscv/boot/dts/renesas/rzfive-smarc-som.dtsi +++ b/arch/riscv/boot/dts/renesas/rzfive-smarc-som.dtsi @@ -22,6 +22,7 @@ &dmac { status = "disabled"; }; +#if (!SW_ET0_EN_N) ð0 { status = "disabled"; @@ -30,6 +31,7 @@ phy0: ethernet-phy@7 { /delete-property/ interrupts; }; }; +#endif ð1 { status = "disabled";
dtbs_check w/ W=1 complains: Warning (unit_address_vs_reg): /soc/ethernet@11c20000/ethernet-phy@7: node has a unit name, but no reg or ranges property Warning (avoid_unnecessary_addr_size): /soc/ethernet@11c20000: unnecessary #address-cells/#size-cells without "ranges" or child "reg" property The ethernet@11c20000 node is guarded by an `#if (!SW_ET0_EN_N)` in rzg2ul-smarc-som.dtsi, where the phy child node is added. In rzfive-smarc-som.dtsi, the ethernet node is marked disabled & the interrupt properties are deleted from the phy child node. As a result, the produced dts looks like: ethernet@11c20000 { compatible = "renesas,r9a07g043-gbeth\0renesas,rzg2l-gbeth"; /* snip */ #address-cells = <0x01>; #size-cells = <0x00>; status = "disabled"; ethernet-phy@7 { }; }; Adding a corresponding `#if (!SW_ET0_EN_N)` around the node in rzfive-smarc-som.dtsi avoids the complaint, as the empty child node is not added: ethernet@11c20000 { compatible = "renesas,r9a07g043-gbeth\0renesas,rzg2l-gbeth"; /* snip */ #address-cells = <0x01>; #size-cells = <0x00>; status = "disabled"; }; Signed-off-by: Conor Dooley <conor.dooley@microchip.com> --- CC: Geert Uytterhoeven <geert+renesas@glider.be> CC: Magnus Damm <magnus.damm@gmail.com> CC: Prabhakar Mahadev Lad <prabhakar.mahadev-lad.rj@bp.renesas.com> CC: Rob Herring <robh+dt@kernel.org> CC: Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org> CC: Conor Dooley <conor+dt@kernel.org> CC: linux-renesas-soc@vger.kernel.org CC: devicetree@vger.kernel.org CC: linux-riscv@lists.infradead.org CC: linux-kernel@vger.kernel.org --- arch/riscv/boot/dts/renesas/rzfive-smarc-som.dtsi | 2 ++ 1 file changed, 2 insertions(+)