Hi Geert, On Wed, Feb 19, 2020 at 04:55:52PM +0100, Geert Uytterhoeven wrote: > On Wed, Feb 19, 2020 at 4:33 PM Laurent Pinchart wrote: > > On Tue, Feb 18, 2020 at 02:30:18PM +0100, Geert Uytterhoeven wrote: > > > Add reset control properties to the device nodes for the Display Units > > > on all supported R-Car Gen3 SoCs. Note that on these SoCs, there is > > > only a single reset for each pair of DU channels. > > > > > > The display nodes on R-Car V3M and V3H already had "resets" properties, > > > but lacked the corresponding "reset-names" properties. > > > > > > Join the clocks lines while at it, to increase uniformity. > > > > > > Signed-off-by: Geert Uytterhoeven <geert+renesas@xxxxxxxxx> > > > > --- a/arch/arm64/boot/dts/renesas/r8a77965.dtsi > > > +++ b/arch/arm64/boot/dts/renesas/r8a77965.dtsi > > > @@ -2503,10 +2503,11 @@ > > > interrupts = <GIC_SPI 256 IRQ_TYPE_LEVEL_HIGH>, > > > <GIC_SPI 268 IRQ_TYPE_LEVEL_HIGH>, > > > <GIC_SPI 270 IRQ_TYPE_LEVEL_HIGH>; > > > - clocks = <&cpg CPG_MOD 724>, > > > - <&cpg CPG_MOD 723>, > > > + clocks = <&cpg CPG_MOD 724>, <&cpg CPG_MOD 723>, > > > <&cpg CPG_MOD 721>; > > > clock-names = "du.0", "du.1", "du.3"; > > > + resets = <&cpg 724>, <&cpg 722>; > > > + reset-names = "du.0", "du.3"; > > > > I wonder if this should be du.2, especially given that 722 corresponds > > to the non-existing DU2 channel. It's a bit of a mess at the hardware > > Just following the bindings: "du.3" is the lowest channel that is affected > by the reset. Yes I was looking at that, and replied to the DT bindings patch. If the outcome of the discussion there is to keep the bindings as-is, you can have my Reviewed-by: Laurent Pinchart <laurent.pinchart@xxxxxxxxxxxxxxxx> for this whole series. > > level :-S > > Note that supporting R-Car H3-N will make your^H^H^H^Hthe rcar-du device > driver writer's life even more miserable, as suddenly there is no longer > a DU2, while the single unified Display Unit node prevents the DTS > writer from not setting the DU2's status to "okay" in the board DTS > file. But you might look at the ports submode? This will require a separate compatible string I'm afraid. -- Regards, Laurent Pinchart
