Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Subject: Re: [ANNOUNCE] KVM Microconference at LPC 2023
Message-ID: <ce7689c9-1c23-68ff-9afb-7d0aeaa9c80a@amazon.es>
Date:   Wed, 7 Jun 2023 14:20:57 +0200
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:102.0)
 Gecko/20100101 Thunderbird/102.11.2
Content-Language: en-US
To:     Paolo Bonzini <pbonzini@redhat.com>,
        KVM list <kvm@vger.kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
CC:     Sean Christopherson <seanjc@google.com>,
        Marc Zyngier <maz@kernel.org>, Alexander Graf <graf@amazon.de>,
        "Cali, Marco" <xmarcalx@amazon.co.uk>
References: <2f19f26e-20e5-8198-294e-27ea665b706f@redhat.com>
 <150e1ad0-3d59-762b-6032-897d5630a3bf@amazon.es>
 <5283b17c-185b-a37f-84e2-dcd08e042ec0@redhat.com>
From:   Babis Chalios <bchalios@amazon.es>
In-Reply-To: <5283b17c-185b-a37f-84e2-dcd08e042ec0@redhat.com>
Content-Type: text/plain; charset="UTF-8"; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk

On 7/6/23 12:37, Paolo Bonzini wrote:
> CAUTION: This email originated from outside of the organization. Do 
> not click links or open attachments unless you can confirm the sender 
> and know the content is safe.
>
>
>
> On 6/7/23 12:13, Babis Chalios wrote:
>>
>>
>> On our side, we 've been working on providing ways to let VMs (kernel
>>  and user space) know that they have been cloned/snapshotted/restored
>> from snapshots[1].
>>
>> This is tightly coupled with PRNGs both in kernel and user space and
>>  there needs to be some collaboration with random.c to tie everything
>> together [2][3]. It sounds like it could be a good fit for this MC
>> (?).
>
> Does the VM generation ID ACPI device from Microsoft work for you?  It
> is not tied to Hyper-V.
>
> Paolo
>

It's a long story :P

TLDR:

For some use-cases it should be enough [1]. For PRNGs though it doesn't 
seem to work as is. (1) The Linux implementation of VMGENID is racy and 
(2) last year there were long discussions related to the user-space
ABI, which diverged us away from it.

BTW, just saw that you have already submitted the proposal for the MC, 
sorry for being late :)

Cheers,
Babis