[PATCH 06/13] arm64: dts: mediatek: mt8188: Update VPPSYS node name and compatible

[Fei Shao]

On Mon, Sep 16, 2024 at 6:02 PM Krzysztof Kozlowski <krzk at kernel.org> wrote:
>
> On 10/09/2024 13:06, Fei Shao wrote:
> > On Tue, Sep 10, 2024 at 3:19 PM Krzysztof Kozlowski <krzk at kernel.org> wrote:
> >>
> >> On 10/09/2024 07:12, Fei Shao wrote:
> >>> On Mon, Sep 9, 2024 at 7:41 PM Krzysztof Kozlowski <krzk at kernel.org> wrote:
> >>>>
> >>>> On 09/09/2024 13:14, Fei Shao wrote:
> >>>>> Use and add "syscon" in VPPSYS node names and compatible to fix errors
> >>>>> from `make CHECK_DTBS=y mediatek/mt8188-evb.dtb`.
> >>>>>
> >>>>> Signed-off-by: Fei Shao <fshao at chromium.org>
> >>>>> ---
> >>>>>
> >>>>>  arch/arm64/boot/dts/mediatek/mt8188.dtsi | 8 ++++----
> >>>>>  1 file changed, 4 insertions(+), 4 deletions(-)
> >>>>>
> >>>>> diff --git a/arch/arm64/boot/dts/mediatek/mt8188.dtsi b/arch/arm64/boot/dts/mediatek/mt8188.dtsi
> >>>>> index 2900d78b7ceb..14e51a11f688 100644
> >>>>> --- a/arch/arm64/boot/dts/mediatek/mt8188.dtsi
> >>>>> +++ b/arch/arm64/boot/dts/mediatek/mt8188.dtsi
> >>>>> @@ -1799,8 +1799,8 @@ mfgcfg: clock-controller at 13fbf000 {
> >>>>>                       #clock-cells = <1>;
> >>>>>               };
> >>>>>
> >>>>> -             vppsys0: clock-controller at 14000000 {
> >>>>> -                     compatible = "mediatek,mt8188-vppsys0";
> >>>>> +             vppsys0: syscon at 14000000 {
> >>>>> +                     compatible = "mediatek,mt8188-vppsys0", "syscon";
> >>>>
> >>>> If this was working before, it looks like this is not a syscon and
> >>>> bindings need to be fixed.
> >>>
> >>> I guess it's because the binding was later updated in commit
> >>> 26bcd8a53098 ("dt-bindings: arm: mediatek: mmsys: Add VPPSYS
> >>> compatible for MT8188"), and the corresponding DT update was unnoticed
> >>> at the time.
> >>> If that makes sense then this should be a valid fix.
> >>
> >> Not necessarily. Why not fixing bindings? Prove that bindings are
> >> correct, not DTS, first.
> >
> > MediaTek's mmsys doesn't merely control clocks, it also provides
> > display pipeline routing control and other misc control registers, so
> > it's appropriate to categorize it as a system controller over a clock
> > controller.
> > As for vdosys and vppsys, they are likely variants or aliases of mmsys
> > introduced in their newer SoCs.
>
> Nothing like that was in the commit msg...

Just for a record, I've revised the commit message in the newer series:
https://lore.kernel.org/all/20240925110044.3678055-7-fshao@chromium.org/

Thanks,
Fei