Hi,

On 05/28/2014 06:43 PM, Mark Brown wrote:

On Wed, May 28, 2014 at 01:47:50PM +0200, Tomasz Figa wrote:

<snip>

quoted

IMHO, all we need here is a way to tell the MMC (or HSIC) core when to
look for a new device and when not (e.g. power down the host controller
completely). Anything else, including proper power sequencing is up to
the platform driver of such non-discoverable device - it's only its host
interface that is discoverable when enabled. I believe this simplistic
approach would lead to much less new code added, better reusability of
code (power sequencing independent of host interface) and no need to
create overly generic code, which usually turns out to be not generic
enough.

We then have the problem of working out where that platform driver comes
from, for many of these buses the device can be completely described as
just a device on the parent bus with some resources connected.

Right, I think what we should do is focus on the DT description for this
first and then worry about the implementation later.

I believe that at the DT level this all should be represented as a child-node
under the host controller. Following that reasoning it makes perfect sense
to just focus on mmc for now, while keeping in mind that it would be nice if
what comes out of this will be re-usable.

I've done a detailed, updated proposal for the mmc case (with admittedly
some implementation comments in there) in my last mail, but unfortunately
I've seen little response to that. Can people please reply to my
proposal where we've 2 levels of child nodes, one per mmc slot, and then the
slot nodes has card / sdio-func child nodes, see my last mail.

Regards,

Hans