On Sat, Jan 13, 2024 at 08:13:35PM +0100, Heiko Stübner wrote:
> Am Samstag, 13. Januar 2024, 19:06:56 CET schrieb Lukas Wunner:
> > Since commit 26c9d152ebf3 ("dt-bindings: tpm: Consolidate TCG TIS
> > bindings"), several issues are reported by "make dtbs_check" for arm64
> > devicetrees:
> >
> > The compatible property needs to contain the chip's name in addition to
> > the generic "tcg,tpm_tis-spi" and the nodename needs to be "tpm@0"
> > rather than "cr50@0":
> >
> > tpm@1: compatible: ['tcg,tpm_tis-spi'] is too short
> > from schema $id: http://devicetree.org/schemas/tpm/tcg,tpm_tis-spi.yaml#
> >
> > cr50@0: $nodename:0: 'cr50@0' does not match '^tpm(@[0-9a-f]+)?$'
> > from schema $id: http://devicetree.org/schemas/tpm/google,cr50.yaml#
> >
> > Fix these schema violations.
[...]
> > arch/arm64/boot/dts/freescale/imx8mm-phygate-tauri-l.dts | 2 +-
> > arch/arm64/boot/dts/freescale/imx8mm-venice-gw72xx.dtsi | 2 +-
> > arch/arm64/boot/dts/freescale/imx8mm-venice-gw73xx.dtsi | 2 +-
> > arch/arm64/boot/dts/freescale/imx8mp-beacon-kit.dts | 2 +-
> > arch/arm64/boot/dts/freescale/imx8mp-venice-gw72xx.dtsi | 2 +-
> > arch/arm64/boot/dts/freescale/imx8mp-venice-gw73xx.dtsi | 2 +-
> > arch/arm64/boot/dts/freescale/imx8mp-venice-gw74xx.dts | 2 +-
> > arch/arm64/boot/dts/freescale/imx8mq-kontron-pitx-imx8m.dts | 2 +-
> > arch/arm64/boot/dts/mediatek/mt8183-kukui.dtsi | 2 +-
> > arch/arm64/boot/dts/mediatek/mt8192-asurada.dtsi | 2 +-
> > arch/arm64/boot/dts/rockchip/rk3399-gru-bob.dts | 2 +-
> > arch/arm64/boot/dts/rockchip/rk3399-gru-scarlet.dtsi | 2 +-
>
> you might want to split this per sub-architecture perhaps (freescale,
> mediatek, rockchip) as such dts changes normally go through the trees
> of the subarchitecture maintainers.
My concern was that other developers may see these new warnings
and start submitting piecemeal fixes. That would be a waste
of time since I already prepared fixes for everything weeks ago.
I was only waiting for the schema changes to hit mainline.
I was hoping that these might make it to Linus during the back half
of the merge window so that the warnings are gone by rc1 time
and nobody will ever see them.
Of course I can split them up if that's what Arnd & Olof prefer.
> For the rockchip-parts itself, I'm also fine with them going through
> somewhere else - gru devices are pretty much "finished" by now,
> so for the rockchip changes
> Acked-by: Heiko Stuebner <heiko@xxxxxxxxx>
Thanks for taking a look!
Lukas
