linux-kernel - Re: [RFC] KVM: x86: Support KVM VMs sharing SEV context

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAKiEG5oicgF=wuYda6RhH_Memc_gnoYSeiimVthkuxckKWN9_w@mail.gmail.com>
Date:   Fri, 5 Mar 2021 12:43:07 -0800
From:   Nathan Tempelman <natet@...gle.com>
To:     Paolo Bonzini <pbonzini@...hat.com>
Cc:     Ashish Kalra <ashish.kalra@....com>,
        Thomas Lendacky <thomas.lendacky@....com>, x86@...nel.org,
        kvm@...r.kernel.org, linux-kernel@...r.kernel.org,
        Steve Rutherford <srutherford@...gle.com>,
        Sean Christopherson <seanjc@...gle.com>,
        David Rientjes <rientjes@...gle.com>,
        Brijesh Singh <brijesh.singh@....com>
Subject: Re: [RFC] KVM: x86: Support KVM VMs sharing SEV context

On Fri, Mar 5, 2021 at 7:13 AM Paolo Bonzini <pbonzini@...hat.com> wrote:
>
> On 05/03/21 15:04, Ashish Kalra wrote:
> >> +    /* Mirrors of mirrors should work, but let's not get silly */
> >> +    if (is_mirroring_enc_context(kvm)) {
> >> +            ret = -ENOTTY;
> >> +            goto failed;
> >> +    }
> > How will A->B->C->... type of live migration work if mirrors of
> > mirrors are not supported ?
>
> Each host would only run one main VM and one mirror, wouldn't it?

That's correct. You could create a second mirror vm of the original
(A->B, A->C) if you needed two in-guest workers, but I don't see a use
for a chain. If anyone can see one I can write it that way, but in the interest
of keeping it simple I've blocked it. Originally I'd built it with
that functionality,
but allowing a chain like that smells like recursion and from what I
understand we don't like recursion in the kernel. There's also the fear as
steve mentioned that we could blow the callstack with a long chain of
destroys starting from the leaf.
Ideally we give userspace one less gun to shoot itself in the foot with.