Re: [PATCH v2 1/3] dt-bindings: clock: add rk3562 cru bindings

["Heiko Stübner" <heiko@sntech.de>]

Am Montag, 24. Februar 2025, 09:52:12 MEZ schrieb Kever Yang:
> Hi Krzysztof,
> 
> On 2024/12/27 16:25, Krzysztof Kozlowski wrote:
> > On Tue, Dec 24, 2024 at 05:23:08PM +0800, Kever Yang wrote:
> >> Document the device tree bindings of the rockchip rk3562 SoC
> >> clock and reset unit.
> >>
> >> Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
> >> ---
> > A nit, subject: drop second/last, redundant "bindings". The
> > "dt-bindings" prefix is already stating that these are bindings.
> > See also:
> > https://elixir.bootlin.com/linux/v6.7-rc8/source/Documentation/devicetree/bindings/submitting-patches.rst#L18
> >
> >
> > s/rk3562/Rocchip RK3562/
> > or whatever your proper name is (and use proper capitalized parts of
> > products)
> Will update.
> >
> >> +properties:
> >> +  compatible:
> >> +    const: rockchip,rk3562-cru
> >> +
> >> +  reg:
> >> +    maxItems: 1
> >> +
> >> +  "#clock-cells":
> >> +    const: 1
> >> +
> >> +  "#reset-cells":
> >> +    const: 1
> >> +
> >> +  clocks:
> >> +    maxItems: 2
> >
> > Why clocks are not required?
> The cru is the clock-controller, which is always on module in SoC,
> so we don't need to enable "clock" for this clock-controller.

hmm, shouldn't clocks be

  clocks:
    minItems: 1
    maxItems: 2

The CRU _needs_ the xin24m because that is the main oscillator
supplying everything, but _can_ work work without xin32k .

Sidenote: itseems we're doing this wrong on rk3588

Heiko