[Xen-ia64-devel] Re: [patch 00/11] ia64: kexec: Map EFI memory in the same location as Linux (v20080320)

[Simon Horman <horms@xxxxxxxxxxxx>]

On Wed, Apr 02, 2008 at 04:55:55PM +0900, Simon Horman wrote:
> On Tue, Mar 25, 2008 at 04:20:56PM +0900, Simon Horman wrote:
> > On Mon, Mar 24, 2008 at 01:05:26PM -0600, Alex Williamson wrote:
> > > 
> > > On Thu, 2008-03-20 at 15:52 +0900, Simon Horman wrote:
> > > > Hi,
> > > > 
> > > > here is another spin of the kexec EFI patches. I think we are getting
> > > > very close if we aren't there already.
> > > 
> > > Hi Simon,
> > > 
> > >    Have you tried launching an HVM domain with this patch set?  I never
> > > get past the white background in the VNC window (before EFI prints OK),
> > > and very shortly after launch, Xen and dom0 die.
> > 
> > No, I haven't tested that. I will look into it.
> > Thanks for the pointer.
> 
> Hi Alex,
> 
> sorry for taking a while to get back to you.
> 
> I've been able to confirm that this is indeed a problem.
> It seems to be related to using the EFI rid in the hypervisor
> for at least SAL. I'm not really sure why this is occuring.
> I'll poke a bit further, but if anyone has any ideas please
> ping me.

I've been able to investigate this problem a little further.
It turns out that things lock up in the first call to 
ia64_new_rr7_efi() via xen_pal_emulator(), which is called
from the firmware of the hvm domain.

I am guessing that (once again) the TLB re-pinning isn't quite right.
I'll poke further and try and fix this.

-- 
Horms


_______________________________________________
Xen-ia64-devel mailing list
Xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-ia64-devel