Oliver, On Tue, Mar 29 2022 at 09:02, Oliver Upton wrote: > On Tue, Mar 29, 2022 at 7:19 AM Thomas Gleixner <tglx@xxxxxxxxxxxxx> wrote: >> > Doing this the other way around (advance the TSC, tell the guest to fix >> > MONOTONIC) is fundamentally wrong, as it violates two invariants of the >> > monotonic clock. Monotonic counts during a migration, which really is a >> > forced suspend. Additionally, you cannot step the monotonic clock. >> >> A migration _should_ have suspend semantics, but the forced suspend >> which is done by migration today does not have proper defined semantics >> at all. >> >> Also clock monotonic can be stepped forward under certain circumstances >> and the kernel is very well able to handle it within well defined >> limits. Think about scheduled out vCPUs. From their perspective clock >> monotonic is stepping forwards. >> > > Right. For better or worse, the kernel has been conditioned to > tolerate small steps due to scheduling. There is just zero definition > around how much slop is allowed.
