Re: [PATCH v3 2/3] dt-bindings: phy: Add constants for lan966x serdes
From: Rob Herring <robh@kernel.org>
Date: 2021-10-19 13:39:55
Also in:
linux-phy, lkml
On Tue, Oct 19, 2021 at 4:11 AM Horatiu Vultur [off-list ref] wrote:
The 10/18/2021 14:28, Rob Herring wrote:quoted
On Fri, Oct 15, 2021 at 02:39:19PM +0200, Horatiu Vultur wrote:quoted
Lan966x has: 2 integrated PHYs, 3 SerDes and 2 RGMII interfaces. Which requires to be muxed based on the HW representation. So add constants for each interface to be able to distinguish them. Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com> --- include/dt-bindings/phy/phy-lan966x-serdes.h | 14 ++++++++++++++ 1 file changed, 14 insertions(+) create mode 100644 include/dt-bindings/phy/phy-lan966x-serdes.hdiff --git a/include/dt-bindings/phy/phy-lan966x-serdes.h b/include/dt-bindings/phy/phy-lan966x-serdes.h new file mode 100644 index 000000000000..8a05f93ecf41 --- /dev/null +++ b/include/dt-bindings/phy/phy-lan966x-serdes.h@@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: (GPL-2.0 OR MIT) */ + +#ifndef __PHY_LAN966X_SERDES_H__ +#define __PHY_LAN966X_SERDES_H__ + +#define PHY(x) (x) +#define PHY_MAX PHY(2) +#define SERDES6G(x) (PHY_MAX + 1 + (x)) +#define SERDES6G_MAX SERDES6G(3) +#define RGMII(x) (SERDES6G_MAX + 1 + (x)) +#define RGMII_MAX RGMII(2) +#define SERDES_MAX (RGMII_MAX + 1)I still don't understand. #phy-cells description says we have: <port idx> <serdes idx> But here it's 3 numbers. How are these defines used to fill in the 2 cells?Actually they are still only a number. Or maybe I am missing something.
So all the defines apply to the 2nd cell? That's what's missing. The cell description needs to spell all this out. 3 different modes or whatever. Explain what the h/w is comprised of in the top level 'description'.
Maybe an example will help:
---
serdes: serdes@e2004010 {
compatible = "microchip,lan966x-serdes";
reg = <0xe202c000 0x9c>, <0xe2004010 0x4>;
#phy-cells = <2>;
};
&port0 {
...
phys = <&serdes 0 SERDES6G(1)>;
...
};
&port1 {
...
phys = <&serdes 1 PHY(0)>;I think CU was better, just needed some comments. PHY is pretty vague.
... } ... --- Here are some existing examples based on which I have created this patch series: https://elixir.bootlin.com/linux/v5.15-rc6/source/arch/mips/boot/dts/mscc/ocelot_pcb120.dts#L99
None of which use PHY() or RGMII()...
https://elixir.bootlin.com/linux/v5.15-rc6/source/arch/mips/boot/dts/mscc/ocelot.dtsi#L274quoted
Rob-- /Horatiu