[ofa-general] Re: InfiniBand card (mthca) in Linux

[Roland Dreier]

 > Well, in Dom0 the action:
 > modprobe ib_mthca
 > rmmod ib_mthca
 > modprobe ib_mthca
 > 
 > kills the machine. However, it is quite strange because it produces oops in
 > XFS (file system), for me, it looks like it does some memory corruption in the
 > kernel and basically I have the same problem in DomU where the same error is
 > induced by the first modprobe ib_mthca.

Loading and unloading ib_mthca many times works fine on a non-Xen
system.  So there is something different about the Xen environment
that is causing a problem.  It could be a bug in mthca exposed by Xen
(eg improper use of of the DMA mapping API or something like that).

Can you turn on all the memory debugging options like SLAB_DEBUG
etc. and see if it turns up anything?

Also I'd be curious to see the exact XFS oops you're getting, since it
might have a clue to what's going on.

 - R.