Re: [PATCH 03/15] KVM: SVM: Inject #UD on RDTSCP when it should be disabled in the guest

Subject: Re: [PATCH 03/15] KVM: SVM: Inject #UD on RDTSCP when it should be disabled in the guest
From: Paolo Bonzini <pbonzini@xxxxxxxxxx>
Date: Tue, 4 May 2021 23:57:27 +0200
Cc: Vitaly Kuznetsov <vkuznets@xxxxxxxxxx>, Wanpeng Li <wanpengli@xxxxxxxxxxx>, Joerg Roedel <joro@xxxxxxxxxx>, kvm list <kvm@xxxxxxxxxxxxxxx>, LKML <linux-kernel@xxxxxxxxxxxxxxx>, Xiaoyao Li <xiaoyao.li@xxxxxxxxx>, Reiji Watanabe <reijiw@xxxxxxxxxx>
In-reply-to: <YJHCadSIQ/cK/RAw@google.com>
References: <20210504171734.1434054-1-seanjc@google.com> <20210504171734.1434054-4-seanjc@google.com> <CALMp9eSvXRJm-KxCGKOkgPO=4wJPBi5wDFLbCCX91UtvGJ1qBg@mail.gmail.com> <YJHCadSIQ/cK/RAw@google.com>
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.8.1

On 04/05/21 23:53, Sean Christopherson wrote:

Does the right thing happen here if the vCPU is in guest mode when
userspace decides to toggle the CPUID.80000001H:EDX.RDTSCP bit on or
off?

I hate our terminology.  By "guest mode", do you mean running the vCPU, or do
you specifically mean running in L2?


Guest mode should mean L2.

(I wonder if we should have a capability that says "KVM_SET_CPUID2 canonly be called prior to KVM_RUN").


Paolo

Follow-Ups:
- Re: [PATCH 03/15] KVM: SVM: Inject #UD on RDTSCP when it should be disabled in the guest
  - From: Jim Mattson

References:
- [PATCH 00/15] KVM: x86: RDPID/RDTSCP fixes and uret MSR cleanups
  - From: Sean Christopherson
- [PATCH 03/15] KVM: SVM: Inject #UD on RDTSCP when it should be disabled in the guest
  - From: Sean Christopherson
- Re: [PATCH 03/15] KVM: SVM: Inject #UD on RDTSCP when it should be disabled in the guest
  - From: Jim Mattson
- Re: [PATCH 03/15] KVM: SVM: Inject #UD on RDTSCP when it should be disabled in the guest
  - From: Sean Christopherson