Re: [PATCH 0/2] RFC: Precise TSC migration

[Paolo Bonzini]

On 01/12/20 20:35, Thomas Gleixner wrote:
> > This makes the random error in calculation of this value invariant
> > across vCPUS, and allows the guest to do kvmclock calculation in userspace
> > (vDSO) since kvmclock parameters are vCPU invariant.
>
> That's not the case today? OMG!

It's optional. If the host tells the guest that the host TSC is messed 
up, kvmclock disables its vDSO implementation and falls back to the syscall.

Paolo