Dave Jones wrote:
Yes the kernel saw some data about the bridges! I do not have a dump to hand but may be able to reproduce if neededOn Wed, Oct 24, 2007 at 06:14:36PM +0100, Howard Wilkinson wrote: On the kernel that works, is anything in the above section different?
pci=routeirq fails with same problem> PCI: If a device doesn't work, try "pci=routeirq". If it helps, post a > report Worth a shot. Also give pci=noacpi a try if that still doesn't fix it.
pci=noacpi works in that it sees the disc but the network will not start the loopback device - probably unrelated as I have seen something similar elsewhere.
This worked and will do me for now. But how do we go about getting this fixed, there are a LARGE number of these boxes in the field!On the face of things, interrupt routing is screwed up somehow, and everything goes south shortly after initialisation. Give those boot params a try, and see if that helps any. Sadly, this isn't going to get fixed in time for F8 now, so we're in the "hope there are workarounds" area for this. (Final big hammer to try if all else fails - acpi=off)