linux-kernel - Re: [PATCH v1 1/3] kvm: svm: Add support for additional SVM NPF error codes

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] [day] [month] [year] [list]

Date:   Tue, 22 Nov 2016 23:29:56 +0100
From:   Paolo Bonzini <pbonzini@...hat.com>
To:     Tom Lendacky <thomas.lendacky@....com>,
        Brijesh Singh <brijesh.singh@....com>, kvm@...r.kernel.org
Cc:     rkrcmar@...hat.com, joro@...tes.org, x86@...nel.org,
        linux-kernel@...r.kernel.org, mingo@...hat.com, hpa@...or.com,
        tglx@...utronix.de, bp@...e.de
Subject: Re: [PATCH v1 1/3] kvm: svm: Add support for additional SVM NPF error
 codes



On 22/11/2016 23:15, Tom Lendacky wrote:
> > 2) what bit is set if the processor is reading the PDPTEs of a 32-bit
> > PAE guest?
>
> I believe that bit 33 will be set.  The PDPE's are considered guest
> tables and are read during a guest table walk (see APM vol2 section
> 15.25.10).  Note that this is slightly different than the bare-metal
> behavior of legacy PAE mode as APM describes. I'll try to test this
> and verify it.

No big deal, indeed it's a bit different from Intel which caches the
four PDPEs, but it's enough to know that bit 33 will be set.

Paolo