Re: [PATCH 6/7] arm64: dts: renesas: r8a779md: Add Renesas R-Car R8A779MD M3Le DTs

[Laurent Pinchart <laurent.pinchart@ideasonboard.com>]

On Sun, May 03, 2026 at 01:39:27AM +0200, Marek Vasut wrote:
> On 4/30/26 12:11 AM, Laurent Pinchart wrote:
> 
> Hello Laurent,
> 
> >> +/delete-node/ &vin4csi20;
> >> +/delete-node/ &vin5csi20;
> >> +/delete-node/ &vin6csi20;
> >> +/delete-node/ &vin7csi20;
> >> +/delete-node/ &xhci0;
> >> +
> >> +&du {
> >> +	compatible = "renesas,du-r8a779md";
> >> +	renesas,cmms = <&cmm0>, <&cmm3>;
> > 
> > Shouldn't you then also delete the cmm1 node ?
> 
> Please see my reply to [PATCH 1/7] dt-bindings: display: renesas,du: 
> Document Renesas R-Car R8A779MD M3Le , the CMM0,1,3 all seems to be on 
> the chip and accessible. I think the CMM driver should be aware of the 
> IP, even if it is unused, e.g. to bring it into low power state if needed.
> 
> > And does the DU still need three clocks and three interrupts ?
> The Geist board schematic indicates that DU_DOTCLKIN1 is present on the 
> SoC package and is supplied by Xtal X22 on the board, so I think the 
> answer is "yes" for the first clock part of your question.
> 
> As for the interrupt part, it seems DU1 is present, but not connected. 
> That means the DU1 interrupt does exist, but is not going to be used. Do 
> you prefer to describe it or leave it out ?

Following our discussions on IRC, I understand there are quite a few
unknows at this point, and more testing is required in order to upstream
display support. I think your next version will drop support for the DU,
I'm fine with that as a first step. We can then build on top.

-- 
Regards,

Laurent Pinchart