Re: [PATCH] libata, dt: Add clocks to sata_rcar bindings

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

Hi Mark,

On Monday 20 January 2014 09:31:19 Mark Rutland wrote:
> On Sun, Jan 19, 2014 at 08:16:42PM +0000, Laurent Pinchart wrote:
> > On Friday 17 January 2014 02:07:42 Valentine Barshak wrote:
> > > Now that the clocks are available in the R-Car Gen2 DT,
> > > add clocks property description to the sata_rcar bindings.
> > > The clocks have been tested on r8a7791 so we use that
> > > as an example of the R-Car SATA node.
> > > 
> > > The patch is against for-next branch of the libata git repo.
> > > 
> > > Signed-off-by: Valentine Barshak <valentine.barshak@cogentembedded.com>
> > > ---
> > > 
> > >  Documentation/devicetree/bindings/ata/sata_rcar.txt | 10 ++++++----
> > >  1 file changed, 6 insertions(+), 4 deletions(-)
> > > 
> > > diff --git a/Documentation/devicetree/bindings/ata/sata_rcar.txt
> > > b/Documentation/devicetree/bindings/ata/sata_rcar.txt index
> > > 1e61113..6da60c0 100644
> > > --- a/Documentation/devicetree/bindings/ata/sata_rcar.txt
> > > +++ b/Documentation/devicetree/bindings/ata/sata_rcar.txt
> > > 
> > > @@ -7,12 +7,14 @@ Required properties:
> > >  			  - "renesas,sata-r8a7791" for R-Car M2
> > >  - reg			: address and length of the SATA registers;
> > >  - interrupts		: must consist of one interrupt specifier.
> > > +- clocks		: must contain a phandle and clock-specifier pair.
> > 
> > I would say "must contain a reference to the functional clock.", as the
> > clock could be referenced by a phandle only depending on the SATA IP core
> > integration in the SoC.
> 
> In that case the clock-specifier is simply zero cells (though admittedly
> a pair including a zero-cells element is a bit odd).
> 
> The wording in the patch is consistent with the form I've been
> recommending elsewhere:
> 
> - clocks: A list of phandles + clock-specifier pairs, one for each
>   entry in clock-names.
> - clock-names: Should contain:
>   * "fclk" - the functional clock
>   * "other_clk" - some other clock.
> 
> > Wouldn't it be time to have standard wordings for clocks (and interrupt)
> > bindings ?
> 
> I would certainly like to see consistent wording across bindings
> (especially for interrupts given the addition of the interrupts-extended
> binding).

What about creating a new file in Documentation/devicetree/bindings/ with 
reference wordings for common properties ?

-- 
Regards,

Laurent Pinchart