Re: [RFC PATCH 1/2] dt-bindings: mtd: Add Cavium SOCs NAND bindings

[Jan Glauber <jan.glauber@caviumnetworks.com>]

On Mon, Apr 03, 2017 at 09:47:05AM -0500, Rob Herring wrote:
> On Mon, Apr 3, 2017 at 9:38 AM, Jan Glauber
> <jan.glauber@caviumnetworks.com> wrote:
> > On Mon, Apr 03, 2017 at 08:29:37AM -0500, Rob Herring wrote:
> >> On Mon, Mar 27, 2017 at 06:05:23PM +0200, Jan Glauber wrote:
> >> > Add device tree binding description for Cavium SOC nand flash controller.
> >> >
> >> > CC: Rob Herring <robh+dt@kernel.org>
> >> > CC: Mark Rutland <mark.rutland@arm.com>
> >> > CC: devicetree@vger.kernel.org
> >> >
> >> > Signed-off-by: Jan Glauber <jglauber@cavium.com>
> >> > ---
> >> >  .../devicetree/bindings/mtd/cavium_nand.txt        | 32 ++++++++++++++++++++++
> >> >  1 file changed, 32 insertions(+)
> >> >  create mode 100644 Documentation/devicetree/bindings/mtd/cavium_nand.txt
> >> >
> >> > diff --git a/Documentation/devicetree/bindings/mtd/cavium_nand.txt b/Documentation/devicetree/bindings/mtd/cavium_nand.txt
> >> > new file mode 100644
> >> > index 0000000..4698d1f
> >> > --- /dev/null
> >> > +++ b/Documentation/devicetree/bindings/mtd/cavium_nand.txt
> >> > @@ -0,0 +1,32 @@
> >> > +* Cavium NAND controller
> >> > +
> >> > +Required properties:
> >> > +
> >> > +- compatible:              should be "cavium,cn8xxx-nand"
> >>
> >> Don't use wildcards in compatible strings. For PCI devices, this should
> >> be based on the PCI vendor and device IDs.
> >>
> >
> > Is there a syntax for compatible PCI devices? I'm afraid I've not seen
> > this yet, can you give an example?
> 
> www.o3one.org/hwdocs/openfirmware/pci_supplement_2_1.pdf

Thanks, I probably should have read this before...

So it will be something like:
compatible = "pci177d,a04f"

A bit unreadable, but it solves the wildcard issue.

> > Most of Cavium's devices are PCI devices, we just added the compatible
> > as convenience and usually it is not parsed.
> 
> Linux doesn't parse it, but it's still required in the binding.

OK.

--Jan

> Rob