RE: [PATCH 03/15] dt-bindings: memory: fsl: convert ifc binding to yaml schema

[Leo Li <leoyang.li@nxp.com>]

> -----Original Message-----
> From: Rob Herring <robh@kernel.org>
> Sent: Monday, February 8, 2021 12:21 PM
> To: Leo Li <leoyang.li@nxp.com>
> Cc: linux-arm-kernel@lists.infradead.org; Oleksij Rempel <linux@rempel-
> privat.de>; Krzysztof Kozlowski <krzk@kernel.org>;
> devicetree@vger.kernel.org; linux-kernel@vger.kernel.org; Shawn Guo
> <shawnguo@kernel.org>
> Subject: Re: [PATCH 03/15] dt-bindings: memory: fsl: convert ifc binding to
> yaml schema
> 
> On Mon, Feb 08, 2021 at 05:07:14PM +0000, Leo Li wrote:
> >
> >
> > > -----Original Message-----
> > > From: Rob Herring <robh@kernel.org>
> > > Sent: Monday, February 8, 2021 10:55 AM
> > > To: Leo Li <leoyang.li@nxp.com>
> > > Cc: linux-arm-kernel@lists.infradead.org; Oleksij Rempel
> > > <linux@rempel- privat.de>; Rob Herring <robh+dt@kernel.org>;
> > > Krzysztof Kozlowski <krzk@kernel.org>; devicetree@vger.kernel.org;
> > > linux- kernel@vger.kernel.org; Shawn Guo <shawnguo@kernel.org>
> > > Subject: Re: [PATCH 03/15] dt-bindings: memory: fsl: convert ifc
> > > binding to yaml schema
> > >
> > > On Fri, 05 Feb 2021 17:47:22 -0600, Li Yang wrote:
> > > > Convert the txt binding to yaml format and add description.  Also
> > > > updated the recommended node name to ifc-bus to align with the
> > > > simple-bus node name requirements.
> > > >
> > > > Signed-off-by: Li Yang <leoyang.li@nxp.com>
> > > > ---
> > > >  .../bindings/memory-controllers/fsl/ifc.txt   |  82 ----------
> > > >  .../bindings/memory-controllers/fsl/ifc.yaml  | 140
> > > > ++++++++++++++++++
> > > >  2 files changed, 140 insertions(+), 82 deletions(-)  delete mode
> > > > 100644
> > > > Documentation/devicetree/bindings/memory-controllers/fsl/ifc.txt
> > > >  create mode 100644
> > > > Documentation/devicetree/bindings/memory-controllers/fsl/ifc.yaml
> > > >
> > >
> > > My bot found errors running 'make dt_binding_check' on your patch:
> > >
> > > yamllint warnings/errors:
> > >
> > > dtschema/dtc warnings/errors:
> > > Documentation/devicetree/bindings/memory-
> > > controllers/fsl/ifc.example.dts:36.27-49.19: Warning (simple_bus_reg):
> > > /example-0/soc/ifc-bus@ffe1e000/flash@0,0: simple-bus unit address
> > > format error, expected "0"
> > > Documentation/devicetree/bindings/memory-
> > > controllers/fsl/ifc.example.dts:51.27-64.19: Warning (simple_bus_reg):
> > > /example-0/soc/ifc-bus@ffe1e000/flash@1,0: simple-bus unit address
> > > format error, expected "100000000"
> > > Documentation/devicetree/bindings/memory-
> > > controllers/fsl/ifc.example.dts:66.26-71.19: Warning (simple_bus_reg):
> > > /example-0/soc/ifc-bus@ffe1e000/cpld@3,0: simple-bus unit address
> > > format error, expected "300000000"
> >
> > Hi Rob,
> >
> > I saw these warnings, but cannot find a good solution to it.  The
> > first cell in the address is the Chip select, while the second cell in
> > the address is the address offset within the chip select.  It would
> > confusing to combine the two cells of different purposes into a single
> > address as suggested by the warning.  Can we allow the multi-cell
> > address in the node name?
> 
> Drop 'simple-bus'. It's not a simple bus. You have registers that presumably
> have some configuration needed.

That's probably true for just using "simple-bus" as compatible along.  But I see many of the current bindings are defining a more specific compatible string in addition to the "simple-bus" compatible and have their own drivers.  I think this probably meet the statement in the device tree spec? "Bindings may be defined as extensions of other each. For example a new bus type could be defined as an extension of the
simple-bus binding."

Regards,
Leo