On Mon, Jul 15, 2024, at 14:12, Herve Codina wrote: > Hi Arnd, > > On Fri, 12 Jul 2024 16:14:31 +0200 > "Arnd Bergmann" <arnd@xxxxxxxx> wrote: > >> On Fri, Jul 12, 2024, at 15:11, Herve Codina wrote: >> > On Thu, 11 Jul 2024 14:33:26 -0600 Rob Herring <robh@xxxxxxxxxx> wrote: >> >> On Thu, Jul 11, 2024 at 1:08 PM Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> wrote: >> >> >> > > >> >> > > This PCI driver purpose is to instanciate many other drivers using a DT >> >> > > overlay. I think MFD is the right subsystem. >> >> >> >> It is a Multi-function Device, but it doesn't appear to use any of the >> >> MFD subsystem. So maybe drivers/soc/? Another dumping ground, but it >> >> is a driver for an SoC exposed as a PCI device. >> >> >> > >> > In drivers/soc, drivers/soc/microchip/ could be the right place. >> > >> > Conor, are you open to have the PCI LAN966x device driver in >> > drivers/soc/microchip/ ? >> >> That sounds like a much worse fit than drivers/mfd: the code >> here does not actually run on the lan966x soc, it instead runs >> on whatever other machine you happen to plug it into as a >> PCI device. > > Maybe drivers/misc ? That's probably a little better, and there is already drivers/misc/mchp_pci1xxxx in there, which also has some aux devices. Maybe we need a new place and then move both of these and some of the similar devices from drivers/mfd to that, but we don't really have to pick one now. Arnd
