linux-kernel - Re: [PATCH v2 6/6] x86: vmx: virtualize split lock detection

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: <CALCETrVuWY4G9-h+m9XAZDscaRrtaZ-j-F_y3qHfENvMkr1g5Q@mail.gmail.com>
Date:   Mon, 3 Feb 2020 10:52:15 -0800
From:   Andy Lutomirski <luto@...nel.org>
To:     Xiaoyao Li <xiaoyao.li@...el.com>
Cc:     Paolo Bonzini <pbonzini@...hat.com>,
        Sean Christopherson <sean.j.christopherson@...el.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
        X86 ML <x86@...nel.org>, kvm list <kvm@...r.kernel.org>,
        LKML <linux-kernel@...r.kernel.org>,
        David Laight <David.Laight@...lab.com>
Subject: Re: [PATCH v2 6/6] x86: vmx: virtualize split lock detection

On Mon, Feb 3, 2020 at 7:21 AM Xiaoyao Li <xiaoyao.li@...el.com> wrote:
>
> Due to the fact that MSR_TEST_CTRL is per-core scope, i.e., the sibling
> threads in the same physical CPU core share the same MSR, only
> advertising feature split lock detection to guest when SMT is disabled
> or unsupported for simplicitly.
>
> Only when host is sld_off, can guest control the hardware value of
> MSR_TEST_CTL, i.e., KVM loads guest's value into hardware when vcpu is
> running.
>
> The vmx->disable_split_lock_detect can be set to true after unhandled
> split_lock #AC in guest only when host is sld_warn mode. It's for not
> burnning old guest, of course malicous guest can exploit it for DoS
> attack.

Is this actually worthwhile?  This only applies to the host having
sld=off or warn and the host having HT off.  I suspect that
deployments supporting migration will not want to use this, and
multi-tenant deployments won't want to use it for SLD-aware guests doe
to DoS risk.

--Andy