On Tue, Dec 13, 2016 at 01:14:17PM -0000, Thomas Gleixner wrote: > --- a/arch/x86/power/cpu.c > +++ b/arch/x86/power/cpu.c > @@ -256,6 +256,7 @@ static void notrace __restore_processor_ > mtrr_bp_restore(); > perf_restore_debug_store(); > msr_restore_context(ctxt); > + tsc_verify_tsc_adjust(true); > }
Should we do that sooner, as in before calling restore_sched_clock_state() ? Otherwise we recompute the sched_clock deltas vs the wrecked TSC and then fix it up through the ADJUST, wrecking our sched clock again.
