Re: [PATCH 04/11] arm64: dts: qcom: pm7250b: make SID configurable

[Luca Weiss]

On Thu Aug 31, 2023 at 2:27 PM CEST, Dmitry Baryshkov wrote:
> On Thu, 31 Aug 2023 at 14:54, Krzysztof Kozlowski
> <krzysztof.kozlowski@xxxxxxxxxx> wrote:
> >
> > On 31/08/2023 13:33, Dmitry Baryshkov wrote:
> > > On Thu, 31 Aug 2023 at 13:13, Luca Weiss <luca.weiss@xxxxxxxxxxxxx> wrote:
> > >>
> > >> On Wed Aug 30, 2023 at 12:06 PM CEST, Krzysztof Kozlowski wrote:
> > >>> On 30/08/2023 11:58, Luca Weiss wrote:
> > >>>> Like other Qualcomm PMICs the PM7250B can be used on different addresses
> > >>>> on the SPMI bus. Use similar defines like the PMK8350 to make this
> > >>>> possible.
> > >>>>
> > >>>> Signed-off-by: Luca Weiss <luca.weiss@xxxxxxxxxxxxx>
> > >>>> ---
> > >>>>  arch/arm64/boot/dts/qcom/pm7250b.dtsi | 23 ++++++++++++++++-------
> > >>>>  1 file changed, 16 insertions(+), 7 deletions(-)
> > >>>>
> > >>>> diff --git a/arch/arm64/boot/dts/qcom/pm7250b.dtsi b/arch/arm64/boot/dts/qcom/pm7250b.dtsi
> > >>>> index e8540c36bd99..3514de536baa 100644
> > >>>> --- a/arch/arm64/boot/dts/qcom/pm7250b.dtsi
> > >>>> +++ b/arch/arm64/boot/dts/qcom/pm7250b.dtsi
> > >>>> @@ -7,6 +7,15 @@
> > >>>>  #include <dt-bindings/interrupt-controller/irq.h>
> > >>>>  #include <dt-bindings/spmi/spmi.h>
> > >>>>
> > >>>> +/* This PMIC can be configured to be at different SIDs */
> > >>>> +#ifndef PM7250B_SID
> > >>>> +   #define PM7250B_SID 2
> > >>>> +#endif
> > >>>
> > >>> Why do you send the same patch as v1, without any reference to previous
> > >>> discussions?
> > >>>
> > >>> You got here feedback already.
> > >>>
> > >>> https://lore.kernel.org/linux-arm-msm/f52524da-719b-790f-ad2c-0c3f313d9fe9@xxxxxxxxxx/
> > >>
> > >> Hi Krzysztof,
> > >>
> > >> I did mention that original patch in the cover letter of this series.
> > >> I'm definitely aware of the discussion earlier this year there but also
> > >> tried to get an update lately if there's any update with no response.
> > >
> > > I think the overall consensus was that my proposal is too complicated
> > > for the DT files.
> >
> > I proposed to duplicate the entries. Do you keep QUP nodes in DTSI and
> > customize per address? No.
>
> At the same time, we do keep SoC files separate from the board files.
> Yes, I'm slightly exaggerating here.
>
> I think that for PMIC files it makes sense to extract common parts if
> that eases reuse of the common parts.

Hi all,

what can I do for v2 now?

1. Keep this patch as-is, and keep pm7250b in device dts.

2. Drop pm7250b patch and drop from device dts, until _someone_ figures
out a solution talking to the PMIC on different SID.

3. Something else like copy-pasting pm7250b.dtsi to pm7250-8.dtsi and
changing the SID there, and using that in device dts.

Please let me know what to do.

Regards
Luca

>
> >
> > I definitely do not agree to these ifndef->define. Maybe using just
> > define would work (so drop ifndef->define), because this makes it
> > obvious and fail-safe if included in wrong place... except that it is
> > still not the define we expect. This is not the coding style present in
> > other DTSes.
> >
> > The true problem how these SPMI bindings were created. Requiring SID
> > address in every child is clearly redundant and I think we do not follow
> > such approach anywhere else.
> >
> > Best regards,
> > Krzysztof
> >