Re: PROBLEM: Devices behind PCI Express-to-PCI bridge not mapped

From: Andreas Koch
Date: Fri Jun 03 2005 - 21:27:25 EST


On Fri, Jun 03, 2005 at 06:55:40PM -0700, Linus Torvalds wrote:
>
>
> On Sat, 4 Jun 2005, Andreas Koch wrote:
> >
> > As you suspected, it wasn't a panacea: The kernel now panics, with a
> > call chain of
> >
> > ...
> > pcibios_init()
> > pci_assign_unassigned_resources()
> > pci_bus_assign_resources()
> > pci_setup_bridge()
> >
> > I can collect more specific info if necessary.
>
> It would be nice to know exactly what it is that panics, I could well
> imagine that it's something like the "bus->self" that ends up being NULL
> for the root bus or something silly like that, simply because x86 has
> never needed to use these functions.
>
> If so, it migth be as easy as just skipping buses that don't have bridge
> device associated with them, but this would require that you try to debug
> the oops a bit to figure out where it is..

Actually, I tried that already. But I didn't get any usable info from
the oops and GDB (`list *pci_setup_bridge+0x1a2' shows an include file,
not a line in the function) . I'll make another attempt tomorrow when
I am more awake :-)

Andreas
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/