[U-Boot] [PATCH V4 6/7] mx6q: mx6qsabrelite: Provide default serial flash bus and chip-select

[Marek Vasut <marek.vasut@gmail.com>]

> On 01/29/2012 01:11 PM, Marek Vasut wrote:
> >> On 01/29/2012 12:18 PM, Marek Vasut wrote:
> >>>> Signed-off-by: Eric Nelson<eric.nelson@boundarydevices.com>
> >>>> Acked-by: Dirk Behme<dirk.behme@de.bosch.com>
> >>>> Acked-by: Stefano Babic<sbabic@denx.de>
> >>>> ---
> >>>> 
> >>>>    include/configs/mx6qsabrelite.h |    3 +++
> >>>>    1 files changed, 3 insertions(+), 0 deletions(-)
> >>>> 
> >>>> diff --git a/include/configs/mx6qsabrelite.h
> >>>> b/include/configs/mx6qsabrelite.h index 8dd6e39..cc770e2 100644
> >>>> --- a/include/configs/mx6qsabrelite.h
> >>>> +++ b/include/configs/mx6qsabrelite.h
> >>>> @@ -46,9 +46,12 @@
> >>>> 
> >>>>    #define CONFIG_CMD_SF
> >>>>    #ifdef CONFIG_CMD_SF
> >>>> 
> >>>> +#define GPIO_3_19 ((2*32)+19)
> >>> 
> >>> I'd expect this to be in platform headers?
> >> 
> >> This is a choice made in the SabreLite design. I don't think
> >> the same choice has been made for other i.MX6 boards.
> > 
> > I mean the definition of the GPIO_3_19 ...
> 
> I don't think we want to set precedent for defining
> constants for the 100s of GPIO numbers.
> 
> That said, I could achieve my objective of clarifying
> what the number meant (that the constant refers to a GP) by
> changing this:
> 
> 	#define CONFIG_SF_DEFAULT_CS   (0|(GPIO_3_19<<8))
> 
> to this
> 
> 	#define CONFIG_SF_DEFAULT_CS   (0|(IMX_GPIO_NR(3,19)<<8))

Why the (0| part ? Anyway, that indeed looks better, more standard even.

And I think for MX5, there was even stuff defining the GPIO numbers imported 
from Linux.

M

> 
> There's a bit of an issue with this. The IMX_GPIO_NR() macro
> is defined in arch-mx6/gpio.h which is normally included after
> mx6qsabrelite.h because the latter defines the machine.

And the CPP will choke on that ?