Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Subject: Re: [PATCH 02/15] KVM: x86: Emulate RDPID only if RDTSCP is supported
To:     Reiji Watanabe <reijiw@google.com>,
        Sean Christopherson <seanjc@google.com>
Cc:     Vitaly Kuznetsov <vkuznets@redhat.com>,
        Wanpeng Li <wanpengli@tencent.com>,
        Jim Mattson <jmattson@google.com>,
        Joerg Roedel <joro@8bytes.org>, kvm@vger.kernel.org,
        linux-kernel@vger.kernel.org, Xiaoyao Li <xiaoyao.li@intel.com>
References: <20210504171734.1434054-1-seanjc@google.com>
 <20210504171734.1434054-3-seanjc@google.com>
 <CAAeT=FycnR2BonmiHSWobsLCGTuQuTS3kg9x_eYCKLRQGOvYzQ@mail.gmail.com>
From:   Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <92131802-68b7-3d27-3e61-9b388bfcbc7f@redhat.com>
Date:   Wed, 5 May 2021 10:01:54 +0200
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
 Thunderbird/78.8.1
MIME-Version: 1.0
In-Reply-To: <CAAeT=FycnR2BonmiHSWobsLCGTuQuTS3kg9x_eYCKLRQGOvYzQ@mail.gmail.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 05/05/21 05:51, Reiji Watanabe wrote:
>> --- a/arch/x86/kvm/cpuid.c
>> +++ b/arch/x86/kvm/cpuid.c
>> @@ -637,7 +637,8 @@ static int __do_cpuid_func_emulated(struct kvm_cpuid_array *array, u32 func)
>>          case 7:
>>                  entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
>>                  entry->eax = 0;
>> -               entry->ecx = F(RDPID);
>> +               if (kvm_cpu_cap_has(X86_FEATURE_RDTSCP))
>> +                       entry->ecx = F(RDPID);
>>                  ++array->nent;
>>          default:
>>                  break;
> I'm wondering if entry->ecx should be set to F(RDPID) here
> even if the CPU supports RDPID natively.
> (i.e. kvm_cpu_cap_has(X86_FEATURE_RDPID) is true)
> 
> The document "Documentation/virt/kvm/api.rst" says:
> ---
> 4.88 KVM_GET_EMULATED_CPUID
> ---------------------------
> <...>
> Userspace can use the information returned by this ioctl to query
> which features are emulated by kvm instead of being present natively.
> ---

Setting it always is consistent with the treatment of MOVBE above. 
Either way is okay but it should be done for both bits.

Paolo