On Mon, Jun 11, 2012 at 09:56:14AM -0600, Stephen Warren wrote: > * Each regulator-id/regulator-compatible value identifies a specific > individual regulator within the chip that contains it. There is only one > of each named regulator, since that's what exists in HW. So, this is > about configuring HW that we know exists (because it's part of the HW > represented by the parent node for the chip) rather than defining which > HW is present on unprobeable busses, as the device-level compatible does. > Given those differences, I really think that using "compatible" in the > name of the property is just going to cause confusion. This doesn't seem terribly different to me especially in some of the cases people want to use this for where we try to describe the subfunctions of the chip using this mechanism. You have a fixed set of regulators that might exist in a superset device (possibly with some incompatibilities, or with additional properties providing more data on the hardware) and then you mix and match what's in the system based on the nodes you register for the subset device you're using. This sort of thing is actually much more idiomatic with DT than it is with platform data (look at how people want to put device nodes in for the MFD subfunctions all the time...). Really all compatible is saying to me is "here's how you understand this, handle it like an X" and this feels exactly the same. Of course, if someone could just fix the DT to actually be able to do key/value pairs, or allow us to do something useful with the "regulator" string we need to put in there...
Attachment:
signature.asc
Description: Digital signature
