Re: [Xen-devel] [PATCH v4 14/14] xen/x86: setup PVHv2 Dom0 ACPI tables

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 21.12.16 at 17:32, <roger.pau@xxxxxxxxxx> wrote:
> On Mon, Dec 12, 2016 at 06:56:36AM -0700, Jan Beulich wrote:
>> >>> On 30.11.16 at 17:49, <roger.pau@xxxxxxxxxx> wrote:
>> > +static int __init hvm_setup_acpi_madt(struct domain *d, paddr_t *addr)
>> > +{
>> > +    struct acpi_table_madt *madt;
>> > +    struct acpi_table_header *table;
>> > +    struct acpi_madt_io_apic *io_apic;
>> > +    struct acpi_madt_local_apic *local_apic;
>> 
>> I think I had asked about the absence of struct acpi_madt_local_x2apic
>> here before, but now that I look again I also wonder how you get away
>> without struct acpi_madt_nmi_source and struct acpi_madt_local_apic_nmi;
> 
> Since we are currently limited to 128 vCPUs (max APIC ID 255), I don't think
> acpi_madt_local_x2apic is required in this scenario:
> 
> https://lists.xen.org/archives/html/xen-devel/2016-11/msg02815.html 
> 
> I will work on adding those entries once that limit is lifted, would you be
> fine with me adding a comment here regarding the no-need of
> acpi_madt_local_x2apic until support for > 128vCPUs is added?

I'm not convinced these table entries are tied to >255 CPUs - I'm
seeing them on systems with far less. Hence I simply wonder what
functionality we may miss to offer to OSes with these tables
absent.

> Regarding the local/IO APIC NMI structures, won't those NMI's be delivered to
> Xen, or are those then re-injected to the guest?

PV Dom0 may get to see NMIs, so I'd expect PVH to behave
similarly.

>> I can kind of see struct acpi_madt_local_apic_override not being
> 
> Since Xen is the one that sets the local APIC address in the MADT, and it
> always matches the position of the emulated local APIC I don't see why we 
> would
> want to use acpi_madt_local_apic_override. I see that your worries are related
> to what would happen if AML tries to modify the MADT, but wouldn't in that 
> case
> modify the native MADT, instead of the crafted one?

Exactly, so how would you see the modification to get
propagated?

Jan


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
https://lists.xen.org/xen-devel