Re: [PATCH/RFC v2 06/11] mmc: tmio-mmc: define device-tree bindings

[Chris Ball <cjb@laptop.org>]

Hi,

On Wed, Jan 30 2013, Guennadi Liakhovetski wrote:
>> On Thu, Jan 24 2013, Guennadi Liakhovetski wrote:
>> > I tried to keep this binding similar to others, that I proposed in "mmc: 
>> > add DT bindings for more MMC capability flags." Actually, the above is 
>> > indeed wrong, I would call it "cap-sdio-irq." And in that patch I tried to 
>> > keep binding names resembling as closely as possible respective MMC_CAP_* 
>> > flags. I think, it would have been better if "enable-sdio-wakeup" and 
>> > "keep-power-in-suspend" were also named, following the same rule, but it's 
>> > too late now. Anyway, I'm not too concerned about the names. We can use 
>> > "enable-sdio-irq" too if you like.
>> 
>> I see.  Okay, let's go with your proposed cap-* for each MMC_CAP_*, and
>> the pm_caps can stay as they are.
>
> Thanks, let's do that. But in fact, in a recent discussion it has been 
> pointed out to me, that this property
>
> +- toshiba,mmc-cap-sdio-irq	: SDIO IRQ signalling should be used, if
> +	supported by the hardware, i.e. set MMC_CAP_SDIO_IRQ if
> +	TMIO_MMC_SDIO_IRQ is also set
>
> should be common for all MMC drivers: it should be possible to decide per 
> SD interface, whether SDIO IRQ signalling should be enabled. What do you 
> think? Shall we add a global "cap-sdio-irq" DT property instead of a 
> toshiba-specific one?

Yes, please.

- Chris.
-- 
Chris Ball   <cjb@laptop.org>   <http://printf.net/>
One Laptop Per Child