linux-kernel - Re: [PATCH 1/2] KVM: x86: Add emulation support for #GP triggered by VM instructions

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: <51f96502-95d2-569c-2973-eb839f84019f@redhat.com>
Date:   Wed, 13 Jan 2021 13:35:02 +0100
From:   Paolo Bonzini <pbonzini@...hat.com>
To:     Sean Christopherson <seanjc@...gle.com>
Cc:     Wei Huang <wei.huang2@....com>, kvm@...r.kernel.org,
        linux-kernel@...r.kernel.org, vkuznets@...hat.com, joro@...tes.org,
        bp@...en8.de, tglx@...utronix.de, mingo@...hat.com, x86@...nel.org,
        jmattson@...gle.com, wanpengli@...cent.com, bsd@...hat.com,
        dgilbert@...hat.com, mlevitsk@...hat.com
Subject: Re: [PATCH 1/2] KVM: x86: Add emulation support for #GP triggered by
 VM instructions

On 12/01/21 18:42, Sean Christopherson wrote:
> On a related topic, it feels like nested should be disabled by default on SVM
> until it's truly ready for primetime, with the patch tagged for stable.  That
> way we don't have to worry about crafting non-trivial fixes (like this one) to
> make them backport-friendly.

Well, that's historical; I wish it had been disabled by default back in 
the day.

However, after 10 years and after the shakedown last year, it's hard to 
justify breaking backwards compatibility.  Nested SVM is not any less 
ready than nested VMX---just a little less optimized for things such as 
TLB flushes and ASID/VPID---even without this fix.  The erratum has 
visible effects only on a minority of AMD systems (it depends on an 
unlucky placement of TSEG on L0), and it is easy to work around it by 
lowering the amount of <4G memory in L1.

Paolo