On Tue, Feb 21, 2006 at 06:02:15PM +0100, Quentin Garnier wrote:
> On Tue, Feb 21, 2006 at 05:40:16PM +0100, Peter Postma wrote:
> > On Tue, Feb 21, 2006 at 05:09:45PM +0100, Quentin Garnier wrote:
> > > One thing I don't quite get in your patch though is the extra layer for
> > > each objects.  hpckbd should attach directly to j720ssp;  the split in
> > > different source files is fine, though.
> > > 
> > > Your j720{kbd,lcd,tp} are only avatars of j720ssp, it doesn't make sense
> > > to simulate devices at this point.
> > > 
> > 
> > I've added the extra layer because the SSP port is part of the SA-1110
> > and the keyboard, touch-panel, etc are connected to that port. To me,
> > it maked sense to seperate it (the hpcsh port also does similar things).
> 
> The reason why hpcsh has j6x0* devices is because they attach to
> agnostic devices which are generic for sh3 ports.  They're kind of
> place holders, collecting the resources and distributing them to all
> their potential children, whoever they are.  j720ssp, on the other hand,
> is already specific to the Jornada 720 machine, so it doesn't have to be
> agnostic.
> 

I see. However, we might want to move j720ssp.c to arm/sa11x0/sa11x0_ssp.c
some day, and then it is also a generic device for arm ports. Only the
j720ssp_readwrite() function might be Jornada 720 specific (or not -- I
couldn't find a description about how this is supposed to work).

-- 
Peter Postma