----- Original Message -----
> From: "Thomas Gleixner" <t...@linutronix.de>
> To: "Paolo Bonzini" <pbonz...@redhat.com>
> Cc: "Peter Zijlstra" <pet...@infradead.org>, "Wanpeng Li"
> <kernel...@gmail.com>, "Mike Galbraith" <efa...@gmx.de>,
> "LKML" <linux-kernel@vger.kernel.org>, "Ingo Molnar" <mi...@elte.hu>,
> "Borislav Petkov" <b...@alien8.de>
> Sent: Monday, February 27, 2017 6:40:46 PM
> Subject: Re: tip.today - scheduler bam boom crash (cpu hotplug)
>
> On Mon, 27 Feb 2017, Paolo Bonzini wrote:
> > diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c
> > index 2724dc82f992..3080b6877190 100644
> > --- a/arch/x86/kernel/tsc.c
> > +++ b/arch/x86/kernel/tsc.c
> > @@ -1398,6 +1398,9 @@ void __init tsc_init(void)
> >
> > use_tsc_delay();
> >
> > + if (!boot_cpu_has(X86_FEATURE_NONSTOP_TSC))
> > + mark_tsc_unstable("not invariant");
>
> Errm, no.
>
> That makes TSC unusable for systems which do not go into C/P states in
> which the TSC stops. There is a world outside KVM ....
Actually I was surprised too by Peter's patch, as it was adding
mark_tsc_unstable pretty much everywhere that didn't have nonstop TSC.
But hopefully it would still be okay to call clear_sched_clock_stable
in tsc_init, in the same way.
Paolo
