[PATCH qemu 11/38] aspeed/smc: get the number of flash modules from hw strapping

[Cédric Le Goater]

> Hmm. I don't know how I feel about mixing the pinmux with the SoC code.
>  If you've seen the kernel pinmux driver this is going to get large, so
> I'd prefer to isolate it a separate file so we can ignore it for the
> most part. It's hard to do that if it's in the SoC code.
> 
> But!
> 
> On the basis of your patch below I'm reconsidering the pinmux idea. The
> problem patch 11/38 was solving was knowing how many flash slaves we
> have. Previously we thought that was completely described by the second
> watchdog bit in the SCU, 
>
> but now I can't think why we aren't simply describing this as a property 
> of the board (say in AspeedBoardConfig) and avoiding pinmux contortions. 
> Is there a problem with that? 

No problem at all. It does not kill puppies.

I was trying to find a way to detect the presence of a second flash module
from the HW configuration. It is still currently a mystery to me because
enabling a function does not mean there is something behind the pins.

> It is yet another property to configure, but it's a board level problem 
> at the end of the day and will be *much* less code.
> 
> Thoughts?

Let's go that way. It should be simple as there is already a "num-cs"
property.

Thanks,

C.