Re: [PATCH v4 5/8] dt-bindings: i2c: realtek,rtl9301-i2c: extend for clocks and RTL9607C support

[Rustam Adilov]

Hello,
On 2026-03-23 19:41, Conor Dooley wrote:
> On Mon, Mar 23, 2026 at 12:13:34PM +0500, Rustam Adilov wrote:
>> Add the "realtek,rtl9607-i2c" compatible for i2c controller on the
>> RTL9607C SoC series.
>> 
>> Add a clocks property to the properties to describe the i2c reference
>> clock and make it available for all the compatibles. This i2c reference
> 
> Why? I can see that you're doing that, and it's good to know that that
> is intentional, but why is it being done? Need to explain that all of
> these devices actually do have this clock etc.

As far as available information goes, there are some datasheets that note
the existence of these clocks. For example, RTL9310 Developer Guide 1.0 in [1]
mentions "I2C master support 12 peripherals with 2 clocks" at page 19.

And RTL9607 calculates the clk_div with this equation found in Realtek SDK code

data = (62500 / clock) - 1;

where clock is clock frequency in kHz and so the 62500 must be the clock frequency
of the i2c master controller clock in kHz. And since there are 2 i2c controllers,
there are 2 clocks.

That is also the reason why it the rtl9607 is the only one to require the clocks to
be specified and also why the clocks property was added at all.

That's how the rtl9607 i2c controller would look like

i2cclock: i2cclock {
	#clock-cells = <0>;
	compatible = "fixed-clock";

	/* 62.5 MHz */
	clock-frequency = <62500000>;
};

i2c@b4 {
	compatible = "realtek,rtl9607-i2c";
	reg = <0xb4 0x4>;
	#address-cells = <1>;
	#size-cells = <0>;

	clocks = <&i2cclock>;
	realtek,scl = <0>;

	i2c@0 {
		reg = <0>;
		#address-cells = <1>;
		#size-cells = <0>;
	};
};

I hope this explains the whole intention behind it. And if there is better way to explain
this in the commit message, let me know.

[1] - https://github.com/plappermaul/realtek-doc/tree/main/datasheets

Best,
Rustam