Re: [PATCH v1 0/2] rsnd: dts: change to use extended audio dmac register

[Simon Horman]

On Fri, Feb 22, 2019 at 12:05:02PM +0900, jiada wrote:
> Hi Geert
> 
> 
> On 2019/02/20 17:10, Kuninori Morimoto wrote:
> > 
> > Hi Geert
> > 
> > > > According to user reference manual for R-CAR H3 and M3-W SoCs,
> > > > in order to access busif4 ~ busif7, extended audio dmac registers
> > > > (PDMASAREn, PDMADAREn, PDMACHCREn)
> > > > need to be used, rather than basic audio dmac registers
> > > > (PDMASARn, PDMADARn, PDMACHCRn)
> > > > 
> > > > This patch set updates H3 (= r8a7795) and M3-W (= r8a7796)
> > > > to use extended audio dmac registers
> > > 
> > > The same change should be applied for M3-N and RZ/G2M, right?
> > > 
> Currently only H3 and M3-W support to use busif other than busif0,
> I feel currently it's only necessary to use extended audio dmac register for
> these SoCs
> 
> > > R-Car E3 and RZ/G2E already use the extended register set, as they do not
> > > have the basic set.
> > > 
> > > For SoCs having both, this feels a bit like describing software policy, instead
> > > of hardware, to me. Would it make sense to extend the audio bindings, and
> > > allow describing both the basic and extended register sets, and let the driver
> > > make the decision which one to use?
> > 
> > I don't think we need to use basic register.
> > This means extended register only is very enough.
> > I'm not sure why datasheet is indicating basic...
> > 
> Extended audio dmac registers covers all function of basic ones,
> so there is no need to use basic registers

This still sounds a lot like a policy decision that should
be made by the driver based on a fuller description of the hardware
in DT.