On Tue, 2008-11-11 at 07:32 -0700, Matthew Wilcox wrote: > > Ahh, ok, lack of pci_dev representing the root host controller > > was indeed the sparc64 case I was talking about. > > I think this is true for most architectures -- certainly powerpc, parisc > and hp's ia64 machines don't have their root bridge in PCI config space. > HP's chipset people called it non-materialised. It really depends on the chipset... We do on some machines, not others, and we certainly don't have a nice way to know, when it's there, which pci_dev on the root bus it is (just gotta assume there's one and only one of class host bridge but that's not even always the case afaik). Anyway, it's orthogonal to my initial question which is, any objection if I make the legacy_io/mem files appear in all busses ? Cheers, Ben. -- To unsubscribe from this list: send the line "unsubscribe linux-pci" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
