Re: cleaning up NMI handling in trap()

[Manuel Bouyer <bouyer%antioche.eu.org@localhost>]

On Fri, Mar 18, 2011 at 05:02:29PM -0500, David Young wrote:
> On Fri, Mar 18, 2011 at 09:50:10PM +0100, Manuel Bouyer wrote:
> > The
> >     printf ("NMI ... going to debugger\n");
> > 
> > is usefull in knowing what's going on. Can we please keep it in, in
> > one form or another ?
> 
> Sure.  I can move it into kdb_trap() and kgdb_trap().

yes please

> 
> > nmi_dispatch() is not available for xen; but it's now called inconditionally
> > from trap(). Did you forget to add a dummy implementation for XEN ?
> 
> I added it to the xen kernel build.  Is there some reason it will not
> work on xen?

I guess I missed this from your patch. Well, it'll probably never be called on
Xen, as the hypervisor will catch it and will probably panic itself.

-- 
Manuel Bouyer <bouyer%antioche.eu.org@localhost>
     NetBSD: 26 ans d'experience feront toujours la difference
--