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;
Date:   Tue, 24 Apr 2018 09:44:15 -0400
User-Agent: K-9 Mail for Android
In-Reply-To: <588983c3-4ceb-071e-260b-869613655951@redhat.com>
References: <24cd527d-5287-f0be-ffe8-eab341bf1d94@redhat.com> <3866d359-0ef8-6a99-6254-84890be62b93@redhat.com> <20180226122205.GG4377@pd.tnic> <fb705c01-9bdd-9924-00c6-70d8ee940fb4@redhat.com> <CANRm+CyyL2jjjn7nZtLJvP+t1pN9ndAGhijmbX4LJh15=O+dvQ@mail.gmail.com> <20180417202417.GA29865@localhost.localdomain> <CANRm+CwA22VmpA5cfQg8mF_Xke_hXtP+_ui5xU+LV8v0JNj=3Q@mail.gmail.com> <20180418090329.GJ29865@localhost.localdomain> <de059c45-d706-ff70-9063-92ac0a807e26@redhat.com> <CANRm+CzxG5cm8ECopzGYpF3E3RTjvFGD3_7o8S=tffFxnvYoiA@mail.gmail.com> <20180424031400.GA22608@char.us.oracle.com> <588983c3-4ceb-071e-260b-869613655951@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain;
 charset=utf-8
Content-Transfer-Encoding: quoted-printable
Subject: Re: [PATCH] KVM: X86: Allow userspace to define the microcode version
To:     Paolo Bonzini <pbonzini@redhat.com>,
        Wanpeng Li <kernellwp@gmail.com>
CC:     Eduardo Habkost <ehabkost@redhat.com>,
        Borislav Petkov <bp@alien8.de>,
        LKML <linux-kernel@vger.kernel.org>, kvm <kvm@vger.kernel.org>,
        =?UTF-8?B?UmFkaW0gS3LEjW3DocWZ?= <rkrcmar@redhat.com>
From:   Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Message-ID: <2944C5CC-A3AF-4B33-B03F-7EC28FC351CB@oracle.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On April 24, 2018 1:09:00 AM EDT, Paolo Bonzini <pbonzini@redhat=2Ecom> wro=
te:
>On 24/04/2018 05:14, Konrad Rzeszutek Wilk wrote:
>> You would need to include the microcode version in the migration
>stream=2E
>>=20
>> But this brings another point - what if we want to manifest certain
>> new CPUID bits?
>
>You don't do that across migration=2E  Generally if you want to do live
>migration and you set up the guest to know everything about the host
>(down to the microcode level), you should make sure your host are
>pretty
>much identical=2E

I understand how it ought to be but sadly the cloud vendors have a mix of =
hardware=2E


With the retpoline/IBRS support (like what RH kernel has) you could migrat=
e from Skylake to Broadwell and switching over from IBRS to retpoline would=
 be good=2E

Hence asking about this=2E

>
>Paolo