On Wed, Oct 29, 2014 at 9:29 AM, Jake Oshins <jakeo@xxxxxxxxxxxxx> wrote: > >>I have no objection to specifying that these reads may be quite slow. >>Guests should only use them at boot and if they have some reason to >>distrust their RNG pool. > >>The latter can legitimately happen after various types of suspend or >>after migration (detected by VM Generation ID, for example). > > Just as a point of clarification, the VM Generation ID changes (at least in the Hyper-V implementation) only when the VM may have observed a different future, as when a VM backup is restored, a checkpoint is applied, etc. It does not change during migration, when the VM is suspended or when it is rebooted. I've heard anecdotes from application vendors saying that there is some other hypervisor that actually does change the ID at these moments and they wanted us to us to fix that, until I explained that I only control Hyper-V. > Fair enough. If the VM may indeed have observed a different future, then I would argue that reseeding the RNG is very important -- more so than after a normal migration. If the VM trusts that its other history hasn't been compromised, then merely mixing in a unique value would get most of the benefit. --Andy > -- Jake Oshins > -- Andy Lutomirski AMA Capital Management, LLC -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
