Re: [PATCH anybus v4 6/7] dt-bindings: anybuss-host: document devicetree binding

[Rob Herring <robh@kernel.org>]

On Thu, Nov 29, 2018 at 2:59 PM Sven Van Asbroeck <thesven73@gmail.com> wrote:
>
> Rob,
>
> Eliminating the 'compatible' property for the anybus gives me an interesting
> devicetree problem.
>
> The imx-weim code naively loops through all subnodes, applying timing settings
> to the CS in the first reg entry.
> In the example below, WEIM CS0 is programmed with the settings in
> fsl,weim-cs-timing, because the CS part of reg is 0:
>
>
> weim: weim@21b8000 {
>         compatible = "fsl,imx6q-weim";
>         reg = <0x021b8000 0x4000>;
>         clocks = <&clks 196>;
>         #address-cells = <2>;
>         #size-cells = <1>;
>         ranges = <0 0 0x08000000 0x08000000>;
>         fsl,weim-cs-gpr = <&gpr>;
>
>         nor@0,0 {
>                 compatible = "cfi-flash";
>                 reg = <0 0 0x02000000>;
>                 #address-cells = <1>;
>                 #size-cells = <1>;
>                 bank-width = <2>;
>                 fsl,weim-cs-timing = <0x00620081 0x00000001 0x1c022000
>                                 0x0000c000 0x1404a38e 0x00000000>;
>         };
> };
>
> The problem is that my 'anybus controller' hardware spans chip selects.
> It requires access to various memory areas. In the example below, CS1
> will not be programmed with timing settings:
>
> weim {
>         controller@0 {
>                 compatible = "arcx,anybus-controller";
>                 reg = <0 0 0x100>, <0 0x400000 0x800>, <1 0x400000 0x800>;
>                 fsl,weim-cs-timing = ...;
>         };
> };
>
> I've coped with this in the past by creating a 'dummy' subnode, without a
> compatible property. But it's ugly:
>
> weim {
>         controller@0 {
>                 compatible = "arcx,anybus-controller";
>                 reg = <0 0 0x100>, <0 0x400000 0x800>, <1 0x400000 0x800>;
>                 fsl,weim-cs-timing = ...;
>         };
>         dummy@1 { /* this works ! */
>                 reg = <1 0 0>;
>                 fsl,weim-cs-timing = ...;
>         };
> };
>
> Is there a better way? I've looked and looked, but can't find anything that's
> similar.
>
> Or should I extend the imx-weim driver to cope?

Yes, I think it should iterate on child reg properties not just nodes.
Hopefully, you don't need different timing for each CS though.

Rob