Re: [RFC] KVM: mm: fd-based approach for supporting KVM guest private memory

On 9/1/21 9:18 AM, James Bottomley wrote:

quoted

So I think there are literally zero code paths that currently call
try_to_unmap() that will actually work like that on TDX.  If we run
out of memory on a TDX host, we can kill the guest completely and
reclaim all of its memory (which probably also involves killing QEMU
or whatever other user program is in charge), but that's really our
only option.

I think our only option for swap is guest co-operation.

Yes, today that's the only choice.  Both TDX and SEV-SNP can
*theoretically* extend their architectures to enable swap with
uncooperative guests.  But, frankly, nobody has been asking for it.

If you care, please ask Intel and AMD to fix this gap.