On Wed, 2021-09-01 at 18:22 +0200, David Hildenbrand wrote: > On 01.09.21 18:18, James Bottomley wrote: > > On Wed, 2021-09-01 at 08:54 -0700, Andy Lutomirski wrote: > > [...] > > > If you want to swap a page on TDX, you can't. Sorry, go directly > > > to jail, do not collect $200. > > > > Actually, even on SEV-ES you can't either. You can read the > > encrypted page and write it out if you want, but unless you swap it > > back to the exact same physical memory location, the encryption key > > won't work. Since we don't guarantee this for swap, I think swap > > won't actually work for any confidential computing environment. > > > > > 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. We're > > going to have to inflate a balloon or something in the guest and > > have the guest driver do some type of bounce of the page, where it > > becomes an unencrypted page in the guest (so the host can read it > > without the physical address keying of the encryption getting in > > the way) but actually encrypted with a swap transfer key known only > > to the guest. I assume we can use the page acceptance > > infrastructure currently being discussed elsewhere to do swap back > > in as well ... the host provides the guest with the encrypted swap > > page and the guest has to decrypt it and place it in encrypted > > guest memory. > > Ballooning is indeed *the* mechanism to avoid swapping in the > hypervisor and much rather let the guest swap. Shame it requires > trusting a guest, which we, in general, can't. Not to mention other > issues we already do have with ballooning (latency, broken auto- > ballooning, over-inflating, ...). Well not necessarily, but it depends how clever we want to get. If you look over on the OVMF/edk2 list, there's a proposal to do guest migration via a mirror VM that invokes a co-routine embedded in the OVMF binary: https://patchew.org/EDK2/20210818212048.162626-1-tobin@xxxxxxxxxxxxx/ This gives us a page encryption mechanism that's provided by the host but accepted via the guest using attestation, meaning we have a mutually trusted piece of code that can use to extract encrypted pages. It does seem it could be enhanced to do swapping for us as well if that's a road we want to go down? James
