[patch 1/1] iMX51: introduce MX51_GPIO_NR

[s.hauer@pengutronix.de (Sascha Hauer)]

On Thu, Nov 25, 2010 at 09:24:56AM +0100, Sascha Hauer wrote:
> On Wed, Nov 24, 2010 at 06:04:09PM +0100, Arnaud Patard wrote:
> > Fabio Estevam <festevam@gmail.com> writes:
> > 
> > Hi,
> > 
> > > Hi Arnaud,
> > >
> > > 2010/11/22 Arnaud Patard <arnaud.patard@rtp-net.org>:
> > >> Currently, to define a GPIO number, we're using something like :
> > >>
> > >> #define EFIKAMX_PCBID0 ? ? ? ? (2*32 + 16)
> > >>
> > >> to define GPIO 3 16.
> > >>
> > >> This is not really readable and it's error prone imho (note the 3 vs 2).
> > >> So, I'm introducing a new macro to define this in a better way. Now, the
> > >> code sample become :
> > >>
> > >> #define EFIKAMX_PCBID0 ? ? ? ? MX51_GPIO_NR(3, 16)
> > >
> > > Can you rename the macro to MX5x_GPIO_NR instead of MX51_GPIO_NR?
> > >
> > > This way we can also use this macro for MX53 and MX508 when they show
> > > up in mainline.
> > 
> > I've been wondering about to use MX5X instead of MX51 but I kept MX51
> > because I didn't know how the GPIO will work on MX53. If they're
> > compatible, you're right, the name should be MX5X_GPIO_NR and not
> > MX51_GPIO_NR. Assuming you mail means that, I'm going to switch to
> > MX5X_GPIO_NR.
> 
> So far all i.MX use the same gpio numbering and I see no reason why this
> should be changed in future SoCs, so I would go for a IMX_GPIO_NR
> instead of SoC specific variants.

Ok, we are talking about Freescale here, but in this very special case I'm
confident they won't change it. I hope none of the hardware engineers
reads this, I'm sure they are always looking for good ideas to make
the life of the software developers harder.

Sascha

-- 
Pengutronix e.K.                           |                             |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |