Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <849757ed499c4190762a28a1bbf819382abaf22c.camel@redhat.com>
Subject: Re: [PATCH v4 06/25] KVM: x86: hyper-v: Cache
 HYPERV_CPUID_NESTED_FEATURES CPUID leaf
From:   Maxim Levitsky <mlevitsk@redhat.com>
To:     Vitaly Kuznetsov <vkuznets@redhat.com>, kvm@vger.kernel.org,
        Paolo Bonzini <pbonzini@redhat.com>,
        Sean Christopherson <seanjc@google.com>
Cc:     Anirudh Rayabharam <anrayabh@linux.microsoft.com>,
        Wanpeng Li <wanpengli@tencent.com>,
        Jim Mattson <jmattson@google.com>,
        linux-hyperv@vger.kernel.org, linux-kernel@vger.kernel.org
Date:   Thu, 14 Jul 2022 12:59:25 +0300
In-Reply-To: <20220714091327.1085353-7-vkuznets@redhat.com>
References: <20220714091327.1085353-1-vkuznets@redhat.com>
         <20220714091327.1085353-7-vkuznets@redhat.com>
Content-Type: text/plain; charset="UTF-8"
User-Agent: Evolution 3.40.4 (3.40.4-5.fc34) 
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

On Thu, 2022-07-14 at 11:13 +0200, Vitaly Kuznetsov wrote:
> KVM has to check guest visible HYPERV_CPUID_NESTED_FEATURES.EBX CPUID
> leaf to know which Enlightened VMCS definition to use (original or 2022
> update). Cache the leaf along with other Hyper-V CPUID feature leaves
> to make the check quick.
> 
> While on it, wipe the whole 'hv_vcpu->cpuid_cache' with memset() instead
> of having to zero each particular member when the corresponding CPUID entry
> was not found.
> 
> Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
> ---
>  arch/x86/include/asm/kvm_host.h |  2 ++
>  arch/x86/kvm/hyperv.c           | 17 ++++++++---------
>  2 files changed, 10 insertions(+), 9 deletions(-)
> 
> diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
> index de5a149d0971..077ec9cf3169 100644
> --- a/arch/x86/include/asm/kvm_host.h
> +++ b/arch/x86/include/asm/kvm_host.h
> @@ -616,6 +616,8 @@ struct kvm_vcpu_hv {
>                 u32 enlightenments_eax; /* HYPERV_CPUID_ENLIGHTMENT_INFO.EAX */
>                 u32 enlightenments_ebx; /* HYPERV_CPUID_ENLIGHTMENT_INFO.EBX */
>                 u32 syndbg_cap_eax; /* HYPERV_CPUID_SYNDBG_PLATFORM_CAPABILITIES.EAX */
> +               u32 nested_eax; /* HYPERV_CPUID_NESTED_FEATURES.EAX */
> +               u32 nested_ebx; /* HYPERV_CPUID_NESTED_FEATURES.EBX */
>         } cpuid_cache;
>  };
>  
> diff --git a/arch/x86/kvm/hyperv.c b/arch/x86/kvm/hyperv.c
> index e08189211d9a..a8e4944ca110 100644
> --- a/arch/x86/kvm/hyperv.c
> +++ b/arch/x86/kvm/hyperv.c
> @@ -2005,31 +2005,30 @@ void kvm_hv_set_cpuid(struct kvm_vcpu *vcpu)
>  
>         hv_vcpu = to_hv_vcpu(vcpu);
>  
> +       memset(&hv_vcpu->cpuid_cache, 0, sizeof(hv_vcpu->cpuid_cache));
> +
>         entry = kvm_find_cpuid_entry(vcpu, HYPERV_CPUID_FEATURES, 0);
>         if (entry) {
>                 hv_vcpu->cpuid_cache.features_eax = entry->eax;
>                 hv_vcpu->cpuid_cache.features_ebx = entry->ebx;
>                 hv_vcpu->cpuid_cache.features_edx = entry->edx;
> -       } else {
> -               hv_vcpu->cpuid_cache.features_eax = 0;
> -               hv_vcpu->cpuid_cache.features_ebx = 0;
> -               hv_vcpu->cpuid_cache.features_edx = 0;
>         }
>  
>         entry = kvm_find_cpuid_entry(vcpu, HYPERV_CPUID_ENLIGHTMENT_INFO, 0);
>         if (entry) {
>                 hv_vcpu->cpuid_cache.enlightenments_eax = entry->eax;
>                 hv_vcpu->cpuid_cache.enlightenments_ebx = entry->ebx;
> -       } else {
> -               hv_vcpu->cpuid_cache.enlightenments_eax = 0;
> -               hv_vcpu->cpuid_cache.enlightenments_ebx = 0;
>         }
>  
>         entry = kvm_find_cpuid_entry(vcpu, HYPERV_CPUID_SYNDBG_PLATFORM_CAPABILITIES, 0);
>         if (entry)
>                 hv_vcpu->cpuid_cache.syndbg_cap_eax = entry->eax;
> -       else
> -               hv_vcpu->cpuid_cache.syndbg_cap_eax = 0;
> +
> +       entry = kvm_find_cpuid_entry(vcpu, HYPERV_CPUID_NESTED_FEATURES, 0);
> +       if (entry) {
> +               hv_vcpu->cpuid_cache.nested_eax = entry->eax;
> +               hv_vcpu->cpuid_cache.nested_ebx = entry->ebx;
> +       }
>  }
>  
>  int kvm_hv_set_enforce_cpuid(struct kvm_vcpu *vcpu, bool enforce)

Makes sense, looks good.

Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>

Best regards,
	Maxim Levitsky