From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <tglx@linutronix.de>
Received: from p4fea4eb5.dip0.t-ipconnect.de ([79.234.78.181] helo=nanos)
	by Galois.linutronix.de with esmtpsa (TLS1.2:DHE_RSA_AES_256_CBC_SHA256:256)
	(Exim 4.80)
	(envelope-from <tglx@linutronix.de>)
	id 1fN1Y6-0004CN-UQ
	for speck@linutronix.de; Sun, 27 May 2018 21:41:51 +0200
Date: Sun, 27 May 2018 21:41:50 +0200 (CEST)
From: Thomas Gleixner <tglx@linutronix.de>
Subject: Re: L1D-Fault KVM mitigation
In-Reply-To: <alpine.LFD.2.21.999.1805271218550.18753@i7.lan>
Message-ID: <alpine.DEB.2.21.1805272141000.1585@nanos.tec.linutronix.de>
References: <1524563292.8691.38.camel@infradead.org>
 <20180424110445.GU4043@hirez.programming.kicks-ass.net>
 <1527068745.8186.89.camel@infradead.org>
 <20180524094526.GE12198@hirez.programming.kicks-ass.net>
 <alpine.DEB.2.21.1805241201510.1577@nanos.tec.linutronix.de>
 <alpine.DEB.2.21.1805241729520.1577@nanos.tec.linutronix.de>
 <d2029ba2-bdad-5bb9-596d-f22a9bfa5b9a@linux.intel.com>
 <alpine.DEB.2.21.1805251152400.1581@nanos.tec.linutronix.de>
 <20180526204319.GB4486@tassilo.jf.intel.com>
 <alpine.LFD.2.21.999.1805261345410.7864@i7.lan>
 <20180527182550.GC4486@tassilo.jf.intel.com>
 <alpine.LFD.2.21.999.1805271131170.18753@i7.lan>
 <a42a4c62-9d2f-b9a2-a58c-0e1aed3d5fe1@citrix.com>
 <alpine.LFD.2.21.999.1805271218550.18753@i7.lan>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="8323329-593528840-1527450110=:1585"
To: speck@linutronix.de
List-ID: <speck.linutronix.de>

--8323329-593528840-1527450110=:1585
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: QUOTED-PRINTABLE

On Sun, 27 May 2018, speck for Linus Torvalds wrote:
> On Sun, 27 May 2018, speck for Andrew Cooper wrote:
> >=20
> > Xen doesn't ever vcpus enter idle themselves.=C2=A0 We trap HLT/etc and=
 will
> > either schedule another VM, or choose to idle the host if there really
> > is nothing else to do.
>=20
> Right. But then there is never any such thing as "guest idle". There is=
=20
> only "host idle" or "host is doing something else entirely".
>=20
> You _could_ have a "idle polling" mode which is separate from the regular=
=20
> host idle loop, I guess.
>=20
> > I've never come across a plausible usecase for letting non-root mode
> > idle the cores into a low power state.=C2=A0 They simply aren't in a po=
sition
> > to know whether other work needs doing or not.
>=20
> Afaik, it mainly makes sense when there is no actual host OS at all, just=
=20
> the bare-metal hypervisor used for partitioning resources, not scheduling=
=20
> them.

Right. That's what the Jailhouse hypervisor does. It's a zero vmexit setup.

Thanks,

=09tglx
--8323329-593528840-1527450110=:1585--