Re: [PATCH v2 1/3] dt-bindings: power: reset: qcom-pon: Add new compatible PMM8654AU

[Rakesh Kota]

On Tue, Feb 10, 2026 at 09:32:18AM +0100, Krzysztof Kozlowski wrote:
> On 10/02/2026 09:26, Rakesh Kota wrote:
> > On Mon, Feb 09, 2026 at 02:49:24PM +0100, Krzysztof Kozlowski wrote:
> >> On 09/02/2026 14:23, Rakesh Kota wrote:
> >>> Add the compatible string "qcom,pmm8654au-pon" for the PMM8654AU PMIC.
> >>> The PON peripheral on PMM8654AU is compatible with PMK8350, so it is
> >>> documented as a fallback to "qcom,pmk8350-pon".
> >>
> >> Drop everything after ,. Do not explain WHAT you did. We see it.
> >>
> >>>
> >>> While PMM8654AU supports additional registers compared to the baseline,
> >>
> >> full stop.
> >>
> >>> there is currently no active use case for these features. This specific
> >>> compatible string reserves the identifier for future hardware-specific
> >>> handling if required.
> >>
> >> All the rest is irrelevant or even wrong. We do not reserve identifiers.
> >> If you want to reserve something, then I need to reject the patch.
> >>
> > Hi Konrad Dybcio,
> > 
> > It appears that Krzysztof Kozlowski has concerns regarding the
> > compatible string reservation for future use cases, noting that
> > identifiers should not be reserved in this manner.
> 
> So do not reserve identifiers but submit bindings reflecting REAL
> hardware being used.
>
Yes, there is a real hardware difference between the PMK8350 and
PMM865AU PON peripherals. The PMM865AU PON is leveraged from the PMK8350
PON and includes extra features, but those features do not have any
active use cases for now.

If you are okay with the new compatible string, I will send V3 and fix
the commit message suggestions.

> > > So, could you please help ACK my first patch-set instead?
> > https://lore.kernel.org/all/20260122-add_pwrkey_and_resin-v1-1-a9fe87537816@xxxxxxxxxxxxxxxx/
> > 
> 
> First patchset is wrong. You already received comments on that.
> 
> Best regards,
> Krzysztof