On Tue, 2013-12-10 at 07:21 -0800, David Cohen wrote: > Hi, > > [snip] > > >> I don't know anything about SFI GPIOs, how they are defined, where their > >> name comes from, and how they are used so my vision may be incomplete. > >> But AFAICT it all comes down to one of these two scenarios: > >> > >> 1) SFI GPIOs are only used in platform code -> using their pin name is > >> ok, device argument should be assumed to be NULL for their matching > > Let's stick to this. > > > >> 2) SFI GPIOs are also consumed by device drivers -> you need a way to > >> match a (device, con_id) pair to your SFI GPIOs so they can be matched > >> exactly and through the names drivers will request. > > Mostly unlikely we go this way. It would mean we don't need SFI at all. > > I'm currently upstreaming Intel-MID's SFI users. This is a forbidden > situation which I'd change before submit the patch.
I'm sorry I didn't get this. You mean we have to have another mapping, or you agree with system-global GPIOs in SFI case? > > Br, David Cohen -- Andy Shevchenko <andriy.shevche...@linux.intel.com> Intel Finland Oy -- To unsubscribe from this list: send the line "unsubscribe linux-gpio" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
