Re: [Qemu-devel] [RFC 3/3] acpi-build: allocate mcfg for multiple host bridges

[Alex Williamson]

On Tue, 22 May 2018 23:58:30 +0300
"Michael S. Tsirkin" <address@hidden> wrote: 
>
> It's not hard to think of a use-case where >256 devices
> are helpful, for example a nested virt scenario where
> each device is passed on to a different nested guest.
>
> But I think the main feature this is needed for is numa modeling.
> Guests seem to assume a numa node per PCI root, ergo we need more PCI
> roots.

But even if we have NUMA affinity per PCI host bridge, a PCI host
bridge does not necessarily imply a new PCIe domain.  Nearly any Intel
multi-socket system proves this.  Don't get me wrong, I'd like to see
PCIe domain support and I'm surprised edk2 is so far from supporting
it, but other than >256 hotpluggable slots, I'm having trouble coming
up with use cases.  Maybe hotpluggable PCI root hierarchies are easier
with separate domains?  Thanks,

Alex