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
References: <1558082990-7822-1-git-send-email-wanpengli@tencent.com> <7787e0cb-2c46-b5b5-94ea-72c061ea0235@redhat.com>
In-Reply-To: <7787e0cb-2c46-b5b5-94ea-72c061ea0235@redhat.com>
From:   Wanpeng Li <kernellwp@gmail.com>
Date:   Mon, 20 May 2019 19:29:01 +0800
Message-ID: <CANRm+CzD1Sc5bYk9B7yzfXnWkUZwB3cr_86a9x3h9W0ROEoksQ@mail.gmail.com>
Subject: Re: [PATCH 1/4] KVM: x86: Disable intercept for CORE cstate read
To:     Paolo Bonzini <pbonzini@redhat.com>
Cc:     LKML <linux-kernel@vger.kernel.org>, kvm <kvm@vger.kernel.org>,
        =?UTF-8?B?UmFkaW0gS3LEjW3DocWZ?= <rkrcmar@redhat.com>,
        Sean Christopherson <sean.j.christopherson@intel.com>,
        Liran Alon <liran.alon@oracle.com>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, 20 May 2019 at 18:30, Paolo Bonzini <pbonzini@redhat.com> wrote:
>
> On 17/05/19 10:49, Wanpeng Li wrote:
> > From: Wanpeng Li <wanpengli@tencent.com>
> >
> > Allow guest reads CORE cstate when exposing host CPU power management capabilities
> > to the guest. PKG cstate is restricted to avoid a guest to get the whole package
> > information in multi-tenant scenario.
>
> Hmm, I am not sure about this.  I can see why it can be useful to run
> turbostat in the guest, but is it a good idea to share it with the

Yeah.

> guest, since it counts from machine reset rather than from VM reset?

I also saw amazon expose these in their nitro c5 instance.

>
> Maybe it could use a separate bit for KVM_CAP_X86_DISABLE_EXITS?

It could be.

Regards,
Wanpeng Li