Re: [PATCH/RFC 1/2] spi: sh-msiof: Add DT support to DMA setup

[Mark Rutland <mark.rutland@arm.com>]

On Fri, Jun 20, 2014 at 02:18:31PM +0100, Geert Uytterhoeven wrote:
> Hi Mark,
> 
> On Fri, Jun 20, 2014 at 2:52 PM, Mark Rutland <mark.rutland@arm.com> wrote:
> > On Fri, Jun 20, 2014 at 11:20:37AM +0100, Geert Uytterhoeven wrote:
> >> Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
> >> ---
> >> The format of the DMA specifiers depends on the DT bindings for SHDMA,
> >> which are still under development.
> >>
> >>  Documentation/devicetree/bindings/spi/sh-msiof.txt | 17 +++++++++++++--
> >>  drivers/spi/spi-sh-msiof.c                         | 25 ++++++++++++++++------
> >>  2 files changed, 33 insertions(+), 9 deletions(-)
> >>
> >> diff --git a/Documentation/devicetree/bindings/spi/sh-msiof.txt b/Documentation/devicetree/bindings/spi/sh-msiof.txt
> >> index f24baf3b6cc1..fc56e312c0bc 100644
> >> --- a/Documentation/devicetree/bindings/spi/sh-msiof.txt
> >> +++ b/Documentation/devicetree/bindings/spi/sh-msiof.txt
> >> @@ -7,7 +7,13 @@ Required properties:
> >>                        Examples with soctypes are:
> >>                        "renesas,msiof-r8a7790" (R-Car H2)
> >>                        "renesas,msiof-r8a7791" (R-Car M2)
> >> -- reg                  : Offset and length of the register set for the device
> >> +- reg                  : A list of offsets and lengths of the register sets for
> >> +                      the device.
> >> +                      If only one register set is present, it is to be used
> >> +                      by both the CPU and the DMA engine.
> >> +                      If two register sets are present, the first is to be
> >> +                      used by the CPU, and the second is to be used by the
> >> +                      DMA engine.
> >
> > I'm missing something here. I'm we're providing the DMA engines through
> > DMA specifiers below, then why do we need the DMA engine address here?
> > Surely they're separate device nodes?
> 
> It's not the DMA engine address, but the second bank of MSIOF addresses.
> 
> The MSIOF has two (identical) sets of register banks: the first one is to
> be accessed by the CPU, the second one (actually only the Transmit FIFO
> and Receive FIFO Data Registers) is to be accessed by the DMA engine.

Ah, I see. Thanks for the clarification.

> > The code update doesn't seem to do anything with the additional reg
> > entry.
> 
> Indeed, as the second bank of registers is already used by spi-sh-msiof.c
> for the non-DT DMA case. Register banks are resources, so they're
> already present in non-DT platform devices.

Ok. That makes sense then.

Cheers,
Mark.