On 11.03.2025 10:17, Alejandro Vallejo wrote:
> On Tue Mar 11, 2025 at 8:30 AM GMT, Jan Beulich wrote:
>> On 10.03.2025 16:25, Alejandro Vallejo wrote:
>>> Commit cefeffc7e583 marked ACPI tables as NVS in the hvmloader path
>>> because SeaBIOS may otherwise just mark it as RAM. There is, however,
>>> yet another reason to do it even in the PVH path. Xen's incarnation of
>>> AML relies on having access to some ACPI tables (e.g: _STA of Processor
>>> objects relies on reading the processor online bit in its MADT entry)
>>>
>>> This is problematic if the OS tries to reclaim ACPI memory for page
>>> tables as it's needed for runtime and can't be reclaimed after the OSPM
>>> is up and running.
>>>
>>> Fixes: de6d188a519f("hvmloader: flip "ACPI data" to "ACPI NVS" type for
>>> ACPI table region)"
>>> Signed-off-by: Alejandro Vallejo <alejandro.vall...@cloud.com>
>>> ---
>>> I really, really, really dislike this idea of accessing the MADT from
>>> AML.
>>
>> I think this isn't Xen's invention, but something I've seen in various
>> systems' AML.
>
> Do you mean ACPI hotplug? I don't think I've ever seen a real system with
> support for it. I don't suppose you remember any specifically? I'd be quite
> interested to have a look at their ACPI dumps.
Well, what I've seen were systems with firmware provisions for hotplug,
even if e.g. the electric aspects were missing. There was a time, after
all, during which Intel was trying to get OEMs to support this.
Jan
