On Wed, May 02, 2001 at 03:26:43PM -0000, eeh@netbsd.org wrote:
> 
> 	On Wed, May 02, 2001 at 03:09:42PM -0000, eeh@netbsd.org wrote:
> 	> Actually, once again, this sounds like a job for device properties.
> 
> 	What, the X problem, or the pciide problems ?
> 
> Both, actually, once we have in-kernel device properties and a utility to 
> query them in userland.
> 
> 	For pciide I agree; the only problem may be that pciide may be probed/attached
> 	before pcib (which is the one which can give the real information)
> 
> Is the pcib a direct parent of pciide or part of a separate sub-tree?  If
> the former then it will always be attached first.

It's on the same PCI bus (in fact part the the same device in a multi-function
device, for current chipsets):
pchb0 at pci0 dev 0 function 0
pchb0: VIA Technologies VT82C691 (Apollo Pro) Host-PCI (rev. 0x44)
ppb0 at pci0 dev 1 function 0: VIA Technologies VT82C598 (Apollo MVP3) PCI-AGP (
rev. 0x00)
pci1 at ppb0 bus 1
[...]
pcib0 at pci0 dev 7 function 0
pcib0: VIA Technologies VT82C596A (Apollo Pro) PCI-ISA Bridge (rev. 0x22)
pciide0 at pci0 dev 7 function 1: VIA Tech VT82C586A IDE Controller (rev. 0x10)

I fact I said pcib because it's what FreeBSD and linux uses to guess the type
of pciide, but I suspect pchb would work too.

--
Manuel Bouyer, LIP6, Universite Paris VI.           Manuel.Bouyer@lip6.fr
--