[PATCH] spi: dt-bindings: spi-controller: Fix #address-cells for slave mode

[PATCH] spi: dt-bindings: spi-controller: Fix #address-cells for slave mode

[Geert Uytterhoeven]

Currently, the DT bindings for an SPI controller specify that
"#address-cells" must be fixed to one.  However, that applies to an SPI
controller in master mode only.  When running in SPI slave mode,
"#address-cells" should be zero.

Fix this making the value of "#address-cells" dependent on the presence
of "spi-slave".

Fixes: 0a1b929356830257 ("spi: Add YAML schemas for the generic SPI options")
Reported-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh-zM6kxYcvzFBBDgjK7y7TUQ@public.gmane.org>
Signed-off-by: Geert Uytterhoeven <geert+renesas-gXvu3+zWzMSzQB+pC5nmwQ@public.gmane.org>
---
Is this the right way to specify this?  I couldn't find an example, so
this needed some trial-and-error.

As of dtc commit 403cc79f06a135ae ("checks: Update SPI bus check for
'spi-slave'") and Linux commit c2e7075ca8303631 ("scripts/dtc: Update to
upstream version v1.4.7-57-gf267e674d145"), dtc knows about SPI slave.

However, when using "#address-cells = <0>" with W=1:

    Warning (avoid_unnecessary_addr_size): /soc/spi@e6e10000: unnecessary #address-cells/#size-cells without "ranges" or child "reg" property

Just removing #address-cells (using /delete-property/ in the board DTS)
to fix this warning causes:

    Warning (spi_bus_bridge): /soc/spi@e6e10000: incorrect #address-cells for SPI bus

as spi_bus_bridge() uses node_addr_cells(), which defaults to 2 (due to
dtc's ppc64 heritage? But node_size_cells() defaults to 1, not 2?).

How should this be fixed?
Should "#address-cells = <0>" be left out or not?
Should node_{addr,size}_cells() in dtc default to zero?

Thanks!
---
 .../devicetree/bindings/spi/spi-controller.yaml   | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/spi/spi-controller.yaml b/Documentation/devicetree/bindings/spi/spi-controller.yaml
index 1e0ca6ccf64bbd0a..ba2646258b94980e 100644
--- a/Documentation/devicetree/bindings/spi/spi-controller.yaml
+++ b/Documentation/devicetree/bindings/spi/spi-controller.yaml
@@ -20,7 +20,8 @@ properties:
     pattern: "^spi(@.*|-[0-9a-f])*$"
 
   "#address-cells":
-    const: 1
+    minimum: 0
+    maximum: 1
 
   "#size-cells":
     const: 0
@@ -52,6 +53,18 @@ properties:
     description:
       The SPI controller acts as a slave, instead of a master.
 
+if:
+  required:
+    [ spi-slave ]
+then:
+  properties:
+    "#address-cells":
+      const: 0
+else:
+  properties:
+    "#address-cells":
+      const: 1
+
 patternProperties:
   "^slave$":
     type: object
-- 
2.17.1

Re: [PATCH] spi: dt-bindings: spi-controller: Fix #address-cells for slave mode

[Maxime Ripard]

[-- Attachment #1: Type: text/plain, Size: 2652 bytes --]

Hi!

On Thu, Feb 27, 2020 at 02:03:23PM +0100, Geert Uytterhoeven wrote:
> Currently, the DT bindings for an SPI controller specify that
> "#address-cells" must be fixed to one.  However, that applies to an SPI
> controller in master mode only.  When running in SPI slave mode,
> "#address-cells" should be zero.
>
> Fix this making the value of "#address-cells" dependent on the presence
> of "spi-slave".
>
> Fixes: 0a1b929356830257 ("spi: Add YAML schemas for the generic SPI options")
> Reported-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
> Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
> ---
> Is this the right way to specify this?  I couldn't find an example, so
> this needed some trial-and-error.
>
> As of dtc commit 403cc79f06a135ae ("checks: Update SPI bus check for
> 'spi-slave'") and Linux commit c2e7075ca8303631 ("scripts/dtc: Update to
> upstream version v1.4.7-57-gf267e674d145"), dtc knows about SPI slave.
>
> However, when using "#address-cells = <0>" with W=1:
>
>     Warning (avoid_unnecessary_addr_size): /soc/spi@e6e10000: unnecessary #address-cells/#size-cells without "ranges" or child "reg" property
>
> Just removing #address-cells (using /delete-property/ in the board DTS)
> to fix this warning causes:
>
>     Warning (spi_bus_bridge): /soc/spi@e6e10000: incorrect #address-cells for SPI bus
>
> as spi_bus_bridge() uses node_addr_cells(), which defaults to 2 (due to
> dtc's ppc64 heritage? But node_size_cells() defaults to 1, not 2?).
>
> How should this be fixed?
> Should "#address-cells = <0>" be left out or not?
> Should node_{addr,size}_cells() in dtc default to zero?
>
> Thanks!
> ---
>  .../devicetree/bindings/spi/spi-controller.yaml   | 15 ++++++++++++++-
>  1 file changed, 14 insertions(+), 1 deletion(-)
>
> diff --git a/Documentation/devicetree/bindings/spi/spi-controller.yaml b/Documentation/devicetree/bindings/spi/spi-controller.yaml
> index 1e0ca6ccf64bbd0a..ba2646258b94980e 100644
> --- a/Documentation/devicetree/bindings/spi/spi-controller.yaml
> +++ b/Documentation/devicetree/bindings/spi/spi-controller.yaml
> @@ -20,7 +20,8 @@ properties:
>      pattern: "^spi(@.*|-[0-9a-f])*$"
>
>    "#address-cells":
> -    const: 1
> +    minimum: 0
> +    maximum: 1

Nit: This would be better described as an enum: [0, 1]

>    "#size-cells":
>      const: 0
> @@ -52,6 +53,18 @@ properties:
>      description:
>        The SPI controller acts as a slave, instead of a master.
>
> +if:
> +  required:
> +    [ spi-slave ]

Nit: Usually, that notation is when you would do it on the same line,
if you want to go to a new line, it would make more sense to use -

Maxime

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]

Re: [PATCH] spi: dt-bindings: spi-controller: Fix #address-cells for slave mode

[Geert Uytterhoeven]

Hi Maxime,

On Thu, Feb 27, 2020 at 2:09 PM Maxime Ripard <maxime-R63rPqgGiG5yDzI6CaY1VQ@public.gmane.org> wrote:
> On Thu, Feb 27, 2020 at 02:03:23PM +0100, Geert Uytterhoeven wrote:
> > Currently, the DT bindings for an SPI controller specify that
> > "#address-cells" must be fixed to one.  However, that applies to an SPI
> > controller in master mode only.  When running in SPI slave mode,
> > "#address-cells" should be zero.
> >
> > Fix this making the value of "#address-cells" dependent on the presence
> > of "spi-slave".
> >
> > Fixes: 0a1b929356830257 ("spi: Add YAML schemas for the generic SPI options")
> > Reported-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh-zM6kxYcvzFBBDgjK7y7TUQ@public.gmane.org>
> > Signed-off-by: Geert Uytterhoeven <geert+renesas-gXvu3+zWzMSzQB+pC5nmwQ@public.gmane.org>

> > --- a/Documentation/devicetree/bindings/spi/spi-controller.yaml
> > +++ b/Documentation/devicetree/bindings/spi/spi-controller.yaml

> > @@ -52,6 +53,18 @@ properties:
> >      description:
> >        The SPI controller acts as a slave, instead of a master.
> >
> > +if:
> > +  required:
> > +    [ spi-slave ]
>
> Nit: Usually, that notation is when you would do it on the same line,
> if you want to go to a new line, it would make more sense to use -

Sorry, being a poor yaml-by-example programmer, I don't understand what
you mean.  Which part do you refer to by "that notation"?

Thanks!

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert-Td1EMuHUCqxL1ZNQvxDV9g@public.gmane.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: [PATCH] spi: dt-bindings: spi-controller: Fix #address-cells for slave mode

[Maxime Ripard]

[-- Attachment #1: Type: text/plain, Size: 1699 bytes --]

On Thu, Feb 27, 2020 at 02:19:08PM +0100, Geert Uytterhoeven wrote:
> Hi Maxime,
>
> On Thu, Feb 27, 2020 at 2:09 PM Maxime Ripard <maxime-R63rPqgGiG5yDzI6CaY1VQ@public.gmane.org> wrote:
> > On Thu, Feb 27, 2020 at 02:03:23PM +0100, Geert Uytterhoeven wrote:
> > > Currently, the DT bindings for an SPI controller specify that
> > > "#address-cells" must be fixed to one.  However, that applies to an SPI
> > > controller in master mode only.  When running in SPI slave mode,
> > > "#address-cells" should be zero.
> > >
> > > Fix this making the value of "#address-cells" dependent on the presence
> > > of "spi-slave".
> > >
> > > Fixes: 0a1b929356830257 ("spi: Add YAML schemas for the generic SPI options")
> > > Reported-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh-zM6kxYcvzFBBDgjK7y7TUQ@public.gmane.org>
> > > Signed-off-by: Geert Uytterhoeven <geert+renesas-gXvu3+zWzMSzQB+pC5nmwQ@public.gmane.org>
>
> > > --- a/Documentation/devicetree/bindings/spi/spi-controller.yaml
> > > +++ b/Documentation/devicetree/bindings/spi/spi-controller.yaml
>
> > > @@ -52,6 +53,18 @@ properties:
> > >      description:
> > >        The SPI controller acts as a slave, instead of a master.
> > >
> > > +if:
> > > +  required:
> > > +    [ spi-slave ]
> >
> > Nit: Usually, that notation is when you would do it on the same line,
> > if you want to go to a new line, it would make more sense to use -
>
> Sorry, being a poor yaml-by-example programmer, I don't understand what
> you mean.  Which part do you refer to by "that notation"?

I meant that usually the [] notation is used over a single line.

If you want to have a list over multiple lines, usually you would have

required:
  - spi-slave

Maxime

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]

Re: [PATCH] spi: dt-bindings: spi-controller: Fix #address-cells for slave mode

[Geert Uytterhoeven]

Hi Maxime,

On Thu, Feb 27, 2020 at 2:51 PM Maxime Ripard <maxime-R63rPqgGiG5yDzI6CaY1VQ@public.gmane.org> wrote:
> On Thu, Feb 27, 2020 at 02:19:08PM +0100, Geert Uytterhoeven wrote:
> > On Thu, Feb 27, 2020 at 2:09 PM Maxime Ripard <maxime-R63rPqgGiG5yDzI6CaY1VQ@public.gmane.org> wrote:
> > > On Thu, Feb 27, 2020 at 02:03:23PM +0100, Geert Uytterhoeven wrote:
> > > > Currently, the DT bindings for an SPI controller specify that
> > > > "#address-cells" must be fixed to one.  However, that applies to an SPI
> > > > controller in master mode only.  When running in SPI slave mode,
> > > > "#address-cells" should be zero.
> > > >
> > > > Fix this making the value of "#address-cells" dependent on the presence
> > > > of "spi-slave".
> > > >
> > > > Fixes: 0a1b929356830257 ("spi: Add YAML schemas for the generic SPI options")
> > > > Reported-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh-zM6kxYcvzFBBDgjK7y7TUQ@public.gmane.org>
> > > > Signed-off-by: Geert Uytterhoeven <geert+renesas-gXvu3+zWzMSzQB+pC5nmwQ@public.gmane.org>
> >
> > > > --- a/Documentation/devicetree/bindings/spi/spi-controller.yaml
> > > > +++ b/Documentation/devicetree/bindings/spi/spi-controller.yaml
> >
> > > > @@ -52,6 +53,18 @@ properties:
> > > >      description:
> > > >        The SPI controller acts as a slave, instead of a master.
> > > >
> > > > +if:
> > > > +  required:
> > > > +    [ spi-slave ]
> > >
> > > Nit: Usually, that notation is when you would do it on the same line,
> > > if you want to go to a new line, it would make more sense to use -
> >
> > Sorry, being a poor yaml-by-example programmer, I don't understand what
> > you mean.  Which part do you refer to by "that notation"?
>
> I meant that usually the [] notation is used over a single line.
>
> If you want to have a list over multiple lines, usually you would have
>
> required:
>   - spi-slave

IC. So I can also write:

    if:
      required: [ spi-slave ]
    else:
      ...

which is even shorter ;-)

Both seem to work fine.

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert-Td1EMuHUCqxL1ZNQvxDV9g@public.gmane.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: [PATCH] spi: dt-bindings: spi-controller: Fix #address-cells for slave mode

[Maxime Ripard]

[-- Attachment #1: Type: text/plain, Size: 2177 bytes --]

On Thu, Feb 27, 2020 at 03:05:56PM +0100, Geert Uytterhoeven wrote:
> Hi Maxime,
>
> On Thu, Feb 27, 2020 at 2:51 PM Maxime Ripard <maxime@cerno.tech> wrote:
> > On Thu, Feb 27, 2020 at 02:19:08PM +0100, Geert Uytterhoeven wrote:
> > > On Thu, Feb 27, 2020 at 2:09 PM Maxime Ripard <maxime@cerno.tech> wrote:
> > > > On Thu, Feb 27, 2020 at 02:03:23PM +0100, Geert Uytterhoeven wrote:
> > > > > Currently, the DT bindings for an SPI controller specify that
> > > > > "#address-cells" must be fixed to one.  However, that applies to an SPI
> > > > > controller in master mode only.  When running in SPI slave mode,
> > > > > "#address-cells" should be zero.
> > > > >
> > > > > Fix this making the value of "#address-cells" dependent on the presence
> > > > > of "spi-slave".
> > > > >
> > > > > Fixes: 0a1b929356830257 ("spi: Add YAML schemas for the generic SPI options")
> > > > > Reported-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
> > > > > Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
> > >
> > > > > --- a/Documentation/devicetree/bindings/spi/spi-controller.yaml
> > > > > +++ b/Documentation/devicetree/bindings/spi/spi-controller.yaml
> > >
> > > > > @@ -52,6 +53,18 @@ properties:
> > > > >      description:
> > > > >        The SPI controller acts as a slave, instead of a master.
> > > > >
> > > > > +if:
> > > > > +  required:
> > > > > +    [ spi-slave ]
> > > >
> > > > Nit: Usually, that notation is when you would do it on the same line,
> > > > if you want to go to a new line, it would make more sense to use -
> > >
> > > Sorry, being a poor yaml-by-example programmer, I don't understand what
> > > you mean.  Which part do you refer to by "that notation"?
> >
> > I meant that usually the [] notation is used over a single line.
> >
> > If you want to have a list over multiple lines, usually you would have
> >
> > required:
> >   - spi-slave
>
> IC. So I can also write:
>
>     if:
>       required: [ spi-slave ]
>     else:
>       ...
>
> which is even shorter ;-)
>
> Both seem to work fine.

The rest of the file uses the multi-line syntax, so I wouldn't do it
for consistency, but yeah, they are equivalent.

Maxime

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]