On Sat, Dec 27, 2025 at 12:17 PM Krzysztof Kozlowski [off-list ref] wrote:

On Thu, Dec 25, 2025 at 05:24:55PM +0100, Alexandre Belloni wrote:

quoted

On 25/12/2025 09:47:34+0100, Krzysztof Kozlowski wrote:

quoted

On 24/12/2025 15:01, Robert Marko wrote:

quoted

On Wed, Dec 24, 2025 at 2:05 PM Krzysztof Kozlowski [off-list ref] wrote:

quoted

On 24/12/2025 11:30, Robert Marko wrote:

quoted

On Wed, Dec 24, 2025 at 11:21 AM Krzysztof Kozlowski [off-list ref] wrote:

quoted

On Tue, Dec 23, 2025 at 09:16:12PM +0100, Robert Marko wrote:

quoted

Add the required LAN969x clock bindings.

I do not see clock bindings actually here. Where is the actual binding?
Commit msg does not help me at all to understand why you are doing this
without actual required bindings.

I guess it is a bit confusing, there is no schema here, these are the
clock indexes that
reside in dt-bindings and are used by the SoC DTSI.

I understand as not used by drivers? Then no ABI and there is no point
in putting them into bindings.

It is not included by the driver directly, but it requires these exact
indexes to be passed
so its effectively ABI.

How it requires the exact index? In what way? I do not see anything in
the gck driver using/relying on these values. Nothing. Please point me
to the line which directly uses these values.... or how many times I
will need to write this is not ABI?

The index here is the exact id that needs to be set in the PMC_PCR
register and so it is dictated by the hardware.

So not a binding between Linux and DTS.

What would be your suggestion on moving forwarding regarding the clock
HW indexes?

LAN966x does the same with HW clock indexes in the dt-bindings includes.

Regards,
Robert

Best regards,
Krzysztof

-- 
Robert Marko
Staff Embedded Linux Engineer
Sartura d.d.
Lendavska ulica 16a
10000 Zagreb, Croatia
Email: robert.marko@sartura.hr
Web: www.sartura.hr