Re: [PATCH 3/5] dt-bindings: net: Add Loongson-1 DWMAC glue layer
From: Keguang Zhang
Date: Fri Aug 18 2023 - 06:43:40 EST
On Wed, Aug 16, 2023 at 8:54 PM Serge Semin <fancer.lancer@xxxxxxxxx> wrote:
>
> Hi Keguang
>
> On Sat, Aug 12, 2023 at 11:11:33PM +0800, Keguang Zhang wrote:
> > Add devicetree binding document for Loongson-1 DWMAC glue layer.
> >
> > Signed-off-by: Keguang Zhang <keguang.zhang@xxxxxxxxx>
> > ---
> > .../bindings/net/loongson,ls1x-dwmac.yaml | 98 +++++++++++++++++++
> > .../devicetree/bindings/net/snps,dwmac.yaml | 2 +
> > 2 files changed, 100 insertions(+)
> > create mode 100644 Documentation/devicetree/bindings/net/loongson,ls1x-dwmac.yaml
> >
> > diff --git a/Documentation/devicetree/bindings/net/loongson,ls1x-dwmac.yaml b/Documentation/devicetree/bindings/net/loongson,ls1x-dwmac.yaml
> > new file mode 100644
> > index 000000000000..150799460599
> > --- /dev/null
> > +++ b/Documentation/devicetree/bindings/net/loongson,ls1x-dwmac.yaml
> > @@ -0,0 +1,98 @@
> > +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
> > +%YAML 1.2
> > +---
> > +$id: http://devicetree.org/schemas/net/loongson,ls1x-dwmac.yaml#
> > +$schema: http://devicetree.org/meta-schemas/core.yaml#
> > +
>
> > +title: Loongson-1 DWMAC glue layer
>
> DT-schemas describe a device. It has nothing to do with the glue
> driver/layer/whatever.
>
OK. But what about the MODULE_DESCRIPTION in dwmac-loongson1.c
MODULE_DESCRIPTION("Loongson1 DWMAC glue layer");
Should the two parts be aligned with each other?
If not, what's your suggestion then?
> Also I suggest to add a brief device description in the
> "description:" property and add there a brief info regarding the SoCs
> the controllers can be found on, the DW (G)MAC IP-core version the
> ethernet controllers are based on and if possible some data about the
> synthesize parameters: SMA (MDIO-bus), Tx/Rx COE, DMA FIFOs size,
> perfect and hash MAC-filters size, L3L4 frame filters availability,
> PHY interfaces (MII, RMII, RGMII, etc), EEE support, IEEE 1588(-2008)
> Timestamping support, PMT and Wake-up frame support, MAC Management
> counters (MMC).
>
> Note DMA FIFO sizes can be also constrained in the properties
> "rx-fifo-depth" and "tx-fifo-depth"; perfect and hash MAC-filter sizes -
> in "snps,perfect-filter-entries" and "snps,multicast-filter-bins".
>
OK. The description could be added in next version.
> > +
> > +maintainers:
> > + - Keguang Zhang <keguang.zhang@xxxxxxxxx>
> > +
> > +select:
> > + properties:
> > + compatible:
> > + contains:
> > + enum:
> > + - loongson,ls1b-dwmac
> > + - loongson,ls1c-dwmac
> > + required:
> > + - compatible
> > +
> > +properties:
> > + compatible:
> > + items:
> > + - enum:
> > + - loongson,ls1b-dwmac
> > + - loongson,ls1c-dwmac
> > + - const: snps,dwmac-3.50a
> > +
> > + reg:
> > + maxItems: 1
> > +
> > + clocks:
> > + maxItems: 1
> > +
> > + clock-names:
> > + const: stmmaceth
> > +
> > + interrupts:
> > + maxItems: 1
> > +
> > + interrupt-names:
> > + const: macirq
> > +
>
> > + syscon:
> > + $ref: /schemas/types.yaml#/definitions/phandle
> > + description:
> > + Phandle to the syscon containing some extra configurations
> > + including PHY interface mode.
>
> I believe the property is supposed to have a vendor-specific name like
> "loongson,ls1-syscon" or similar.
This has been fixed in v2.
Could you please review v2?
Thanks!
>
> > +
> > +required:
> > + - compatible
> > + - reg
> > + - clocks
> > + - clock-names
> > + - interrupts
> > + - interrupt-names
> > + - phy-handle
>
> > + - phy-mode
>
> You may want to specify the enum-constraints with the value permitted
> for the particular Loongson (G)MAC controller. Seeing ls1b and ls1c
> imply different sets of the PHY-modes the constraints are better to be
> defined in the allOf sub-schemas. Alternatively you can split the
> DT-schema file into two: one for ls1b-dwmac, another one for
> ls1c-dwmac. IMO the later option seems better.
>
The "phy-mode", as pointed by Krzysztof, is defined in
ethernet-controller and already required by snps,dwmac.
So I have dropped it in v2.
For allOf sub-schemas, do you mean something below?
allOf:
- $ref: snps,dwmac.yaml#
- if:
properties:
compatible:
contains:
const: loongson,ls1b-dwmac
then:
properties:
phy-mode:
enum:
- mii
- rgmii
- if:
properties:
compatible:
contains:
const: loongson,ls1c-dwmac
then:
properties:
phy-mode:
enum:
- mii
- rmii
> -Serge(y)
>
> > + - syscon
> > +
> > +allOf:
> > + - $ref: snps,dwmac.yaml#
> > +
> > +unevaluatedProperties: false
> > +
> > +examples:
> > + - |
> > + #include <dt-bindings/clock/loongson,ls1x-clk.h>
> > + #include <dt-bindings/interrupt-controller/irq.h>
> > +
> > + gmac0: ethernet@1fe10000 {
> > + compatible = "loongson,ls1b-dwmac", "snps,dwmac-3.50a";
> > + reg = <0x1fe10000 0x10000>;
> > +
> > + clocks = <&clkc LS1X_CLKID_AHB>;
> > + clock-names = "stmmaceth";
> > +
> > + interrupt-parent = <&intc1>;
> > + interrupts = <2 IRQ_TYPE_LEVEL_HIGH>;
> > + interrupt-names = "macirq";
> > +
> > + phy-handle = <&phy0>;
> > + phy-mode = "mii";
> > +
> > + snps,pbl = <1>;
> > + syscon = <&syscon>;
> > +
> > + mdio {
> > + #address-cells = <1>;
> > + #size-cells = <0>;
> > + compatible = "snps,dwmac-mdio";
> > +
> > + phy0: ethernet-phy@0 {
> > + reg = <0x0>;
> > + };
> > + };
> > + };
> > diff --git a/Documentation/devicetree/bindings/net/snps,dwmac.yaml b/Documentation/devicetree/bindings/net/snps,dwmac.yaml
> > index ddf9522a5dc2..e1a956cf171e 100644
> > --- a/Documentation/devicetree/bindings/net/snps,dwmac.yaml
> > +++ b/Documentation/devicetree/bindings/net/snps,dwmac.yaml
> > @@ -66,6 +66,8 @@ properties:
> > - ingenic,x2000-mac
> > - loongson,ls2k-dwmac
> > - loongson,ls7a-dwmac
> > + - loongson,ls1b-dwmac
> > + - loongson,ls1c-dwmac
> > - qcom,qcs404-ethqos
> > - qcom,sa8775p-ethqos
> > - qcom,sc8280xp-ethqos
> > --
> > 2.39.2
> >
--
Best regards,
Keguang Zhang