RE: [PATCH] Docs: dt: Be explicit and consistent in reference to IOMMU specifiers

[Stuart Yoder]

> -----Original Message-----
> From: Mark Rutland [mailto:mark.rutland@xxxxxxx]
> Sent: Friday, December 16, 2016 5:33 AM
> To: Stuart Yoder <stuart.yoder@xxxxxxx>
> Cc: robh+dt@xxxxxxxxxx; devicetree@xxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; joro@xxxxxxxxxx;
> iommu@xxxxxxxxxxxxxxxxxxxxxxxxxx; will.deacon@xxxxxxx
> Subject: Re: [PATCH] Docs: dt: Be explicit and consistent in reference to IOMMU specifiers
> 
> Hi Stuart,
> 
> On Thu, Dec 15, 2016 at 06:16:13PM -0600, Stuart Yoder wrote:
> > The generic IOMMU binding says that the meaning of an 'IOMMU specifier'
> > is defined by the binding of a specific SMMU.  The ARM SMMU binding
> > never explicitly uses the term 'specifier' at all.  Update implicit
> > references to use the explicit term.
> >
> > In the iommu-map binding change references to iommu-specifier to
> > "IOMMU specifier" so we are 100% consistent everywhere with terminology
> > and capitalization.
> 
> Elsewhere, we always use lower case "xxx-specifier" or "xxx specifier",
> e.g. Documentation/devicetree/bindings/gpio/gpio.txt defines
> "gpio-specifier", ePAPR defines "interrupt specifier".
> 
> Given we're morstly consistent on "iommu-specifier" today,could we
> please jsut update the ARM SMMU binding to match that? If we're going to
> fix the dash mismatch, that's a more general, cross-binding thing.

The notable place where we don't use "iommu-specifier" in in the generic
IOMMU binding itself where we use "IOMMU specifier".  You're suggesting
using "iommu-specifier" everywhere including the generic binding?  Sounds
fine to me.  It's a nit but would like to see it consistent everywhere.

Thanks,
Stuart