Re: [PATCH net-next 08/15] net: dsa: sja1105: transition OF-based MDIO drivers to standalone

[Lee Jones <lee@kernel.org>]

On Thu, 20 Nov 2025, Vladimir Oltean wrote:

> On Thu, Nov 20, 2025 at 04:36:03PM +0000, Lee Jones wrote:
> > The canonical answer goes 3 ways: If you want to use the MFD API, move
> > the handling to drivers/mfd.
> 
> Wait, so now it's negotiable? What about "this device is not an MFD"
> from the other patch? Move it to another folder and suddenly it's an MFD?
> https://lore.kernel.org/netdev/20251120153622.p6sy77coa3de6srw@skbuf/

The "canonical answer" is the general answer, the one that applies to
everyone, not just your use-case.  I'm not (yet?) convinced that this is
an MFD, so at the moment, option one is not on the table for you.

> > If it's possible, use one of the predetermined helpers like
> > of_platform_populate() (and I think we authored another one that
> > worked around some of its issues, but I forget where we put it!).
> 
> yay...
> 
> I don't need the code necessarily, just the overall idea if you remember it.

It was an extension to "simple-bus".  Perhaps "simple-mfd"?

Again, it might not work for you, but it's an avenue worth exploring.

> > Or if all else fails, you have to register the device the old
> > fashioned way with the platform_device_*() API.
> 
> Do you realize that by this, you are inviting me to waste time until the
> next reviewer rightfully asks, why didn't you use MFD, then I'll point
> them towards this discussion, but you didn't really answer?

This is still an abuse of the MFD API.  Used outside of the subsystem
for something that isn't an MFD.  I like the idea of not duplicating
code, but the fact remains that this is a hack.

Another more recent avenue you may explore is the Auxiliary Bus.

-- 
Lee Jones [李琼斯]