Re: [PATCH] rockchip: rk3568-nanopi-r5: Drop duplicated/extra sdhci

Hi Quentin,

> On 12/9/25 10:03 PM, Peter Robinson wrote:
> > The pinctrl settings are now upstream, with the upstream
> > voltage supplies, and the HS200 mode is confirmed to work
> > fine but the HS400 mode had reported issues so let's just
> > consume what the upstream DT has now as it's all known good.
> >
> > Reported-by: Dusty Mabe <[email protected]>
> > Tested-by: Dusty Mabe <[email protected]>
> > Signed-off-by: Peter Robinson <[email protected]>
> > ---
> >   arch/arm/dts/rk3568-nanopi-r5s-u-boot.dtsi | 8 --------
> >   1 file changed, 8 deletions(-)
> >
> > diff --git a/arch/arm/dts/rk3568-nanopi-r5s-u-boot.dtsi b/arch/arm/dts/rk3568-nanopi-r5s-u-boot.dtsi
> > index 64c43374c04..7e24be1a5e7 100644
> > --- a/arch/arm/dts/rk3568-nanopi-r5s-u-boot.dtsi
> > +++ b/arch/arm/dts/rk3568-nanopi-r5s-u-boot.dtsi
> > @@ -12,14 +12,6 @@
> >       /delete-property/ vpcie3v3-supply;
> >   };
> >
> > -&sdhci {
> > -     cap-mmc-highspeed;
>
> I'm assuming this isn't "necessary" since we have HS200 enabled? It
> isn't part of the upstream Linux DTS.

Sorry, I read that as a statement, not a question. No, it's not
necessary and I've got a few users that have been using this patch in
Fedora without issues booting off eMMC for quite some time.

> Considering we have HS200 support enabled in both SPL and proper, it
> should be used, so:

Ultimately if it's needed it should go upstream, but it doesn't appear
to cause issues.

> Reviewed-by: Quentin Schulz <[email protected]>
>
> Thanks!
> Quentin