On 06/11/19 09:32, Peter Zijlstra wrote: >>> The only way virt topology can make any sense what so ever is if the >>> vcpus are pinned to physical CPUs. >> >> This is a subset of the requirements for "trustworthy" SMT. You can have: >> >> - vCPUs pinned to two threads in the same core and exposed as multiple >> cores in the guest > > Why the .... would one do anything like that? If a vCPUs from a different guest could be pinned to a threads in the same core as this guest (e.g. guests with an odd number of vCPUs), then why not. Side-channel wise, you're screwed anyway. >> - vCPUs from different guests pinned to two threads in the same core >> >> and that would be okay as far as KVM_HINTS_REALTIME is concerned, but >> would still allow exploitation of side-channels, respectively within the >> VM and between VMs. > > Hardly, RT really rather would not have SMT. SMT is pretty crap for > determinism. True, but not a problem as long as the guest knows that - it can ignore one sibling for each core for RT tasks, and use hyperthreading for non-RT and housekeeping tasks. Paolo
