Re: ovmf / PCI passthrough impaired due to very limiting PCI64 aperture

[Laszlo Ersek]

On 06/17/20 18:33, Eduardo Habkost wrote:
> On Wed, Jun 17, 2020 at 12:57:52PM -0300, Guilherme Piccoli wrote:
>> Can't qemu reads the host physical bits and pass that as fw_cfg as
>> "real_host_physbits" or something like that?
>> OVMF could rely on that - if such property is available, we use it to
>> extend the PCI64 aperture.
> 
> Giving a exact "real host physbits" value would be too limiting
> if we still want to allow migration to smaller hosts.  But we can
> provide extra information on CPUID or fw_cfg, yes.
> 
> I will try to write down a proposal to extend the KVM CPUID leaf
> to provide extra information about MAXPHYADDR guarantees.

Thank you -- if this can be detected in the guest via CPUID, then I
prefer that to fw_cfg. We need to keep an eye on x-file-slots for fw_cfg.

Thanks
Laszlo