On Thu, 2008-03-27 at 02:29 +0300, Ivan Kokshaysky wrote:

quoted text
> Well, at this point (pdev_sort_resources call) bridge resource->start
> has nothing to do with a bus address, it just a temporary storage for
> required alignment, filled by sizing routines (and 0 is definitely
> invalid here).

Allright, that sounds better then.

quoted text
>  I know, this is quite confusing, but I didn't want to add extra
> fields to existing structures or create temporary per-bus trees...
> But after pci_assign_resource() that resource can certainly be at 0,
> depending on PCIBIOS_MIN_{IO,MEM} and free slots in the resource tree.

In fact, this is the resource assignment code, and PCIBIOS_MIN_* is set
to non-0 for powerpc too so we don't try to assign things down at 0
anyway, so we don't have a problem. We only care about things at 0 that
have already been put there by the firmware and that we decide not to
reassign.

So I don't have any objection anymore.

Cheers,
Ben.


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