Re: [PATCH] ARM: dts: Kill off skeleton{64}.dtsi

[Kevin Hilman <khilman-rdvid1DuHRBWk0Htik3J/w@public.gmane.org>]

Rob Herring <robh-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org> writes:

> Remove the usage of skeleton.dtsi in the remaining dts files. It was
> deprecated since commit 9c0da3cc61f1 ("ARM: dts: explicitly mark
> skeleton.dtsi as deprecated"). This will make adding a unit-address to
> memory nodes easier.
>
> The main tricky part to removing skeleton.dtsi is we could end up with
> no /memory node at all when a bootloader depends on one being present. I
> hacked up dtc to check for this condition.

[...]

I gave this a spin through the kernelCI on multiple SoCs, and builds and
no new regressions for build or boot that I spotted.

Tested-by: Kevin Hilman <khilman-rdvid1DuHRBWk0Htik3J/w@public.gmane.org>

>  arch/arm/boot/dts/meson.dtsi                   |  3 ++-
>  arch/arm/boot/dts/meson6-atv1200.dts           |  1 +
>  arch/arm/boot/dts/meson8-minix-neo-x8.dts      |  1 +
>  arch/arm/boot/dts/meson8b-ec100.dts            |  1 +
>  arch/arm/boot/dts/meson8b-mxq.dts              |  1 +
>  arch/arm/boot/dts/meson8b-odroidc1.dts         |  1 +
>  arch/arm/boot/dts/meson8m2-mxiii-plus.dts      |  1 +

also for the amlogic/meson platforms,

Reviewed-by: Kevin Hilman <khilman-rdvid1DuHRBWk0Htik3J/w@public.gmane.org>
Acked-by: Kevin Hilman <khilman-rdvid1DuHRBWk0Htik3J/w@public.gmane.org>

Kevin