linux-kernel - Re: [PATCH] KVM: SVM: Use a separate vmcb for the nested L2 guest

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]

Date:   Thu, 8 Oct 2020 07:52:46 +0200
From:   Paolo Bonzini <pbonzini@...hat.com>
To:     Maxim Levitsky <mlevitsk@...hat.com>,
        Cathy Avery <cavery@...hat.com>, linux-kernel@...r.kernel.org,
        kvm@...r.kernel.org
Cc:     vkuznets@...hat.com, wei.huang2@....com
Subject: Re: [PATCH] KVM: SVM: Use a separate vmcb for the nested L2 guest

On 08/10/20 00:14, Maxim Levitsky wrote:
>>
>> +	if (svm->vmcb01->control.asid == 0)
>> +		svm->vmcb01->control.asid = svm->nested.vmcb02->control.asid;
> 
> I think that the above should be done always. The asid field is currently host
> controlled only (that is L2 value is ignored, selective ASID tlb flush is not
> advertized to the guest and lnvlpga is emulated as invlpg). 

Yes, in fact I suggested that ASID should be in svm->asid and moved to
svm->vmcb->asid in svm_vcpu_run.  Then there's no need to special case
it in nested code.

This should be a patch coming before this one.

> 
> 1. Something wrong with memory types - like guest is using UC memory for everything.
>     I can't completely rule that out yet

You can print g_pat and see if it is all zeroes.

In general I think it's better to be explicit with vmcb01 vs. vmcb02,
like Cathy did, but I can see it's a matter of personal preference to
some extent.

Paolo