On Fri, Sep 27, 2019 at 07:19:14PM +0200, Paolo Bonzini wrote: > On 27/09/19 19:14, Sean Christopherson wrote: > >> > >> Perhaps we can make all MSRs supported unconditionally if > >> host_initiated. For unsupported performance counters it's easy to make > >> them return 0, and allow setting them to 0, if host_initiated > > I don't think we need to go that far. Allowing any ol' MSR access seems > > like it would cause more problems than it would solve, e.g. userspace > > could completely botch something and never know. > > Well, I didn't mean really _all_ MSRs, only those returned by > KVM_GET_MSR_INDEX_LIST. Ah, that makes way more sense :-) > > For the perf MSRs, could we enumerate all arch perf MSRs that are supported > > by hardware? That would also be the list of MSRs that host_initiated MSR > > accesses can touch regardless of guest support. > > Yes, that is easy indeed. Any ideas about VMX? Can you elaborate on the problem with VMX MSRs?
