[PATCH 1/2] gpio/mxc: get rid of the uses of cpu_is_mx()

[Shawn Guo]

On Mon, Jul 04, 2011 at 11:49:21AM +0200, Sascha Hauer wrote:
> On Mon, Jul 04, 2011 at 05:28:01PM +0800, Shawn Guo wrote:
> > On Mon, Jul 04, 2011 at 08:46:03AM +0200, Sascha Hauer wrote:
> > > On Sun, Jul 03, 2011 at 04:16:56PM +0800, Shawn Guo wrote:
> > > > The patch removes all the uses of cpu_is_mx().  Instead, it utilizes
> > > > platform_device_id to distinguish the gpio differences among SoCs.
> > > > 
> > > > Signed-off-by: Shawn Guo <shawn.guo at linaro.org>
> > > > Cc: Grant Likely <grant.likely at secretlab.ca>
> > > > Cc: Sascha Hauer <s.hauer at pengutronix.de>
> > > > ---
> > > >  arch/arm/mach-imx/mm-imx1.c                   |    8 +-
> > > >  arch/arm/mach-imx/mm-imx21.c                  |   12 +-
> > > >  arch/arm/mach-imx/mm-imx25.c                  |    8 +-
> > > >  arch/arm/mach-imx/mm-imx27.c                  |   12 +-
> > > >  arch/arm/mach-imx/mm-imx31.c                  |    6 +-
> > > >  arch/arm/mach-imx/mm-imx35.c                  |    6 +-
> > > >  arch/arm/mach-mx5/mm-mx50.c                   |   12 +-
> > > >  arch/arm/mach-mx5/mm.c                        |   22 ++--
> > > >  arch/arm/plat-mxc/devices/platform-gpio-mxc.c |    4 +-
> > > >  arch/arm/plat-mxc/include/mach/common.h       |    2 +-
> > > >  drivers/gpio/gpio-mxc.c                       |  123 +++++++++++++++++++++----
> > > >  11 files changed, 151 insertions(+), 64 deletions(-)
> > > 
> > > 
> > > Summarizing the renames up:
> > > 
> > > i.MX1  -> imx1-gpio
> > > i.MX21 -> imx2-gpio
> > > i.MX25 -> imx-gpio
> > > i.MX27 -> imx2-gpio
> > > i.MX31 -> imx-gpio
> > > i.MX35 -> imx-gpio
> > > i.MX50 -> imx-gpio
> > > i.MX51 -> imx-gpio
> > > i.MX53 -> imx-gpio
> > > 
> > > This is not consitent. Please either use the full SoC name for all
> > > device ids or use something like imx-gpio-v1, v2...
> > > It's not good that the i.MX25 is not a imx2 and that the 'modern'
> > > i.MXs only have imx-gpio. We all know that your hardware designers
> > > will be creative enough to change the register layout again in the
> > > future.
> > > 
> > Ok, here it is.  It's avoid confusion in machine code, but every
> > time we add a new soc we need to change touch this table, even if
> > the new soc has a total compatible gpio to IMX_GPIO.  I'm fine with
> > either way.
> 
> I'm also fine with imx1-gpio, imx21-gpio and imx31-gpio and all others
> get a compatible entry with these like you did in the uart driver.
> I only want to avoid to have a imx2-gpio when a i.MX25 is incompatible
> with this.
> When we do so it's probably worth to put this into a comment somewhere
> next to the id table. I can imagine people wonder why only such ancient
> SoCs are supported.
> 
So I'm going to do the following to address your concern.  Please let
me know if you still do not buy it.

static struct platform_device_id mxc_gpio_devtype[] = {
        {
                .name = "imx1-gpio",
                .driver_data = IMX1_GPIO,
        }, {
                .name = "imx21-gpio",
                .driver_data = IMX2_GPIO,
        }, {
                .name = "imx25-gpio",
                .driver_data = IMX_GPIO,
        }, {
                .name = "imx27-gpio",
                .driver_data = IMX2_GPIO,
        }, {
                .name = "imx-gpio",
                .driver_data = IMX_GPIO,
        }, {
                /* sentinel */
        }
};

-- 
Regards,
Shawn