Re: [PATCH] GPIO_SCH and I2C_SCH: select MFD_SUPPORT to eliminate kconfig warnings

[Jean Delvare]

Hi Randy,

On Tue, 14 Dec 2010 16:26:56 -0800, Randy Dunlap wrote:
> From: Randy Dunlap <randy.dunlap@xxxxxxxxxx>
> 
> I2C_ISCH and GPIO_SCH need to select MFD_SUPPORT so that the
> dependency tree hierarchy is satisfied.  This eliminates the
> kconfig warning:
> 
> warning: (I2C_ISCH && I2C && PCI || GPIO_SCH && GPIOLIB && PCI) selects LPC_SCH which has unmet direct dependencies (MFD_SUPPORT && PCI)
> 
> Signed-off-by: Randy Dunlap <randy.dunlap@xxxxxxxxxx>
> Cc: linux-i2c@xxxxxxxxxxxxxxx
> ---
>  drivers/gpio/Kconfig       |    1 +
>  drivers/i2c/busses/Kconfig |    1 +
>  2 files changed, 2 insertions(+)
> 
> --- lnx-2637-rc4.orig/drivers/gpio/Kconfig
> +++ lnx-2637-rc4/drivers/gpio/Kconfig
> @@ -102,6 +102,7 @@ config GPIO_VR41XX
>  config GPIO_SCH
>  	tristate "Intel SCH GPIO"
>  	depends on GPIOLIB && PCI
> +	select MFD_SUPPORT
>  	select MFD_CORE
>  	select LPC_SCH
>  	help
> --- lnx-2637-rc4.orig/drivers/i2c/busses/Kconfig
> +++ lnx-2637-rc4/drivers/i2c/busses/Kconfig
> @@ -107,6 +107,7 @@ config I2C_I801
>  config I2C_ISCH
>  	tristate "Intel SCH SMBus 1.0"
>  	depends on PCI
> +	select MFD_SUPPORT
>  	select MFD_CORE
>  	select LPC_SCH
>  	help

It seems wrong to have to select both MFD_SUPPORT and MFD_CORE. Can't
the MFD subsystem be cleaned up so that subdrivers have a single symbol
to select beyond their own?

I don't quite get the point of MFD_CORE. MFD_SUPPORT could be made
tristate, and MFD_CORE dropped (and then MFD_SUPPORT renamed to
MFD_CORE to minimize the changes.) This is exactly how I2C support is
implemented, for example.

-- 
Jean Delvare
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/