Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
Message-ID: <6b9a1ec2-5ebd-4624-a825-3f31db5cefb5@default>
Date:   Sun, 28 Jan 2018 16:39:48 -0800 (PST)
From:   Liran Alon <liran.alon@oracle.com>
To:     <dwmw2@infradead.org>
Cc:     <konrad.wilk@oracle.com>, <luto@kernel.org>, <tglx@linutronix.de>,
        <torvalds@linux-foundation.org>, <gregkh@linuxfoundation.org>,
        <asit.k.mallick@intel.com>, <dave.hansen@intel.com>,
        <karahmed@amazon.de>, <jun.nakajima@intel.com>,
        <dan.j.williams@intel.com>, <ashok.raj@intel.com>,
        <daniel.kiper@oracle.com>, <arjan.van.de.ven@intel.com>,
        <tim.c.chen@linux.intel.com>, <pbonzini@redhat.com>,
        <linux-kernel@vger.kernel.org>, <ak@linux.intel.com>,
        <kvm@vger.kernel.org>, <aarcange@redhat.com>
Subject: Re: [PATCH] x86: vmx: Allow direct access to MSR_IA32_SPEC_CTRL
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


----- dwmw2@infradead.org wrote:

> On Sun, 2018-01-28 at 15:21 -0500, Konrad Rzeszutek Wilk wrote:
> > >To avoid the overhead of atomically saving and restoring the
> MSR_IA32_SPEC_CTRL
> > >for guests that do not actually use the MSR, only
> add_atomic_switch_msr when a
> > >non-zero is written to it.
> >=20
> >=20
> > We tried this and found that it was about 3% slower that doing the
> > old way of rdmsr and wrmsr.
> >=20
> > But that was also with the host doing IBRS=C2=A0 as well.
>=20
> The common case will be that neither host nor guest are doing IBRS.
> Until the guest touches the MSR we do absolutely *nothing* with it,
> which is definitely the fastest option.

Windows use IBRS and Microsoft don't have any plans to switch to retpoline.
Running a Windows guest should be a pretty common use-case no?

In addition, your handle of the first WRMSR intercept could be different.
It could signal you to start doing the following:
1. Disable intercept on SPEC_CTRL MSR.
2. On VMEntry, Write vCPU SPEC_CTRL value into physical MSR.
3. On VMExit, read physical MSR into vCPU SPEC_CTRL value.
(And if IBRS is used at host, also set physical SPEC_CTRL MSR here to 1)

That way, you will both have fastest option as long as guest don't use IBRS
and also won't have the 3% performance hit compared to Konrad's proposal.

Am I missing something?

-Liran