Re: [PATCH v3] [POWERPC] 85xx: Add basic Uniprocessor MPC8572 DS port

On Sep 13, 2007, at 12:06 PM, Segher Boessenkool wrote:

quoted

quoted

quoted

+		PowerPC,8572@0 {

Maybe it would be good to use "PowerPC,e500" instead -- it would
make it easier to probe for the actual CPU type, that way.  Not
that Linux uses the name/compatible here at all ;-)

I thought about this, not sure what the best solution is.

Since the CPU cores on all these SoCs are identical (well, there
might be a few revisions, or different cache sizes or such -- minor
differences that can be probed for separately), it probably is a
good idea to name them in the tree instead of having each client
have its own table.

Or is there anything about the CPU that can be derived from "8572"
but not from "e500"?

Only in so much that we need something that states what the actual  
processor is.

quoted

quoted

quoted

+	soc8572@ffe00000 {

You should put an interrupt-parent in here, so you can get rid of
it in all the children.

Are interrupt-parent's inherited by child nodes?

A node without "interrupt-parent" uses the regular tree parent for
walking the interrupt "tree".

quoted

quoted

And then there's the pci_bridge thing we're discussing on IRC, of
course -- basically, get rid of the pci_bridge pseudo-node, and
move the interrupt-map for the south-bridge devices into the
south-bridge node.

Leaving the interrupt-map in the PHB because that works and moving  
it down has issues.

Okay, fair enough.  Are you looking at resolving those kernel issues?

No.  I've had enough of this device tree foo for a while :)

[I'm happy to test any patches related to this, if someone else comes  
up with them]

- k