Re: ACPI-Tables corrupted?

["Jan Beulich" <JBeulich@novell.com>]

>>> On 28.07.10 at 14:45, Keir Fraser <keir.fraser@eu.citrix.com> wrote:
> On 28/07/2010 13:13, "Juergen Gross" <juergen.gross@ts.fujitsu.com> wrote:
> 
>>>> As Dom0 is a pv-kernel, it should be able to ignore this entry.
>>>> The crash kernel OTOH should not panic due to the trashed entry!
>>>> What is the correct solution here?
>>> 
>>> Could provide a cmdline option to not nobble the DMAR?
>> 
>> That's a possibility.
>> I wonder whether it wouldn't be better to let dom0 decide not to use it if
>> running under xen. This would remove the requirement for zapping the ACPI
>> table. IMO it's always a bad idea to change data of a deeper layer...
> 
> If we don't zap the DMAR then every existing dom0 kernel will fail with new
> hypervisor.

Who decided that this zapping is going to work on all systems in the
future? E.g. I can't see why a BIOS shouldn't decide to put most of
the ACPI tables in chipset write protected memory, if a chipset offers
such? In such an environment, Dom0 would still see the original
signature - shouldn't we thus correct the original mistake of fiddling
with the ACPI tables as we now have to touch that code anyway?

Jan