On Wed, Jan 05, 2005 at 09:04:01AM -0800, J. William Campbell wrote:
Hi Sven,
Thanks for the comments. I think you are correct that it is at least an
interrupt assignment problem.
Cool.
I will look into this and see if I can figure out what is wrong. I will
compile a Kernel and build an initrd
as soon as I get my cross-development system running. Unfortunately I have
no power-pc based workstations, so it will require developing and building
on an Intel platform. I probably will want to sync up with your
development at that point, so I will have your latest patches for the pci
irq situation. When everything is ready,
Well, i can build kernels for you, just really don't have time to go hunting
for patches.
I will e-mail you to sync up. It probably will be at least 2 weeks because
of other work assignments.
Well, ok.
The MVME boards do use a pci bus internally, just like the MCP750 does,
although the assignments are
somewhat different. The interface to the VME bus is via a bridge (Tundra
Universe) that lives on the pci
bus on one side and the vme bus on the other. The boards therefore are pci
from the kernel point of view,
all VME stuff is inside a driver.
Cool.
BTW, any particular reason why you took this off list ?
Friendly,
Sven Luther