On 16.04.2021 16:35, Thomas Bogendoerfer wrote:
On Fri, Apr 16, 2021 at 12:29:46PM +0300, Sergei Shtylyov wrote:
On 16.04.2021 11:52, Thomas Bogendoerfer wrote:
Add device tree bindings for ethernet controller integrated into
IDT 79RC3243x SoCs.
Signed-off-by: Thomas Bogendoerfer <tsbog...@alpha.franken.de>
---
.../bindings/net/idt,3243x-emac.yaml | 74 +++++++++++++++++++
1 file changed, 74 insertions(+)
create mode 100644 Documentation/devicetree/bindings/net/idt,3243x-emac.yaml
diff --git a/Documentation/devicetree/bindings/net/idt,3243x-emac.yaml
b/Documentation/devicetree/bindings/net/idt,3243x-emac.yaml
new file mode 100644
index 000000000000..3697af5cb66f
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/idt,3243x-emac.yaml
@@ -0,0 +1,74 @@
+# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/idt,3243x-emac.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: IDT 79rc3243x Ethernet controller
+
+description: Ethernet controller integrated into IDT 79RC3243x family SoCs
+
+maintainers:
+ - Thomas Bogendoerfer <tsbog...@alpha.franken.de>
+
+allOf:
+ - $ref: ethernet-controller.yaml#
+
+properties:
+ compatible:
+ const: idt,3243x-emac
+
+ reg:
+ maxItems: 3
+
+ reg-names:
+ items:
+ - const: korina_regs
+ - const: korina_dma_rx
+ - const: korina_dma_tx
+
+ interrupts:
+ items:
+ - description: RX interrupt
+ - description: TX interrupt
+
+ interrupt-names:
+ items:
+ - const: korina_rx
+ - const: korina_tx
+
+ clocks:
+ maxItems: 1
+
+ clock-names:
+ items:
+ - const: mdioclk
+
+required:
+ - compatible
+ - reg
+ - reg-names
+ - interrupts
+ - interrupt-names
+
+additionalProperties: false
+
+examples:
+ - |
+
+ ethernet@60000 {
+ compatible = "idt,3243x-emac";
+
+ reg = <0x60000 0x10000>,
+ <0x40000 0x14>,
+ <0x40014 0x14>;
+ reg-names = "korina_regs",
+ "korina_dma_rx",
+ "korina_dma_tx";
+
+ interrupts-extended = <&rcpic3 0>, <&rcpic3 1>;
You use this prop, yet don't describe it?
that's just interrupt-parent and interrupts in one statement. And since
I know. :-) Yet you don't mention it, making the "interrupts" prop
mandatory instead. And as both interrupt parents are the same here, it does
not even seem justified...
make dt_binding_check didn't complained I thought that's good this way.
Rob, do I need to describe interrupts-extended as well ?
I could change that to interrupt-parent/interrupts as the driver no
longer uses dma under/overrun interrupts, which have a different
interrupt-parent.
In DT, we describe the hardware, not the driver's capabilities.
Thomas.
MBR, Sergei
