On 05/03/2021 15:37, Jan Beulich wrote: > We can't make correctness of our own behavior dependent upon a > hypervisor underneath us correctly telling us the true physical address > with hardware uses. Without knowing this, we can't be certain reserved > bit faults can actually be observed. Therefore, besides evaluating the > number of address bits when deciding whether to use the optimization, > also check whether we're running virtualized ourselves.
I think it would be helpful to point out why we can't even probe at boot - the behaviour may genuinely change as we migrate, and if we ever end up on an IceLake system levelled down for compatibility with older CPUs, the "paddr_bits < PADDR_BITS" check will malfunction in an unsafe way. > > Requested-by: Andrew Cooper <andrew.coop...@citrix.com> > Signed-off-by: Jan Beulich <jbeul...@suse.com> Acked-by: Andrew Cooper <andrew.coop...@citrix.com> This wants backporting to stable releases, so I would recommend for 4.15 even at this point. ~Andrew