Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Subject: Re: [PATCH 3/6] KVM: x86: Add dedicated emulator helper for grabbing
 CPUID.maxphyaddr
To:     Jan Kiszka <jan.kiszka@siemens.com>,
        Sean Christopherson <sean.j.christopherson@intel.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: <20200302195736.24777-1-sean.j.christopherson@intel.com>
 <20200302195736.24777-4-sean.j.christopherson@intel.com>
 <de2ed4e9-409a-6cb1-e295-ea946be11e82@redhat.com>
 <617748ab-0edd-2ccc-e86b-b86b0adf9d3b@siemens.com>
From:   Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <4ddde497-9c71-d64c-df20-3b4439664336@redhat.com>
Date:   Tue, 3 Mar 2020 11:14:22 +0100
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101
 Thunderbird/68.4.1
MIME-Version: 1.0
In-Reply-To: <617748ab-0edd-2ccc-e86b-b86b0adf9d3b@siemens.com>
Content-Type: text/plain; charset=windows-1252
Content-Language: en-US
Content-Transfer-Encoding: 8bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 03/03/20 10:48, Jan Kiszka wrote:
>>
>> I don't think this is a particularly useful change.? Yes, it's not
>> intuitive but is it more than a matter of documentation (and possibly
>> moving the check_cr_write snippet into a separate function)?
> 
> Besides the non obvious return value of the current function, this
> approach also avoids leaving cpuid traces for querying maxphyaddr, which
> is also not very intuitive IMHO.

There are already other cases where we leave CPUID traces.  We can just
stop tracing if check_limit (which should be renamed to from_guest) is
true; there are other internal cases which call ctxt->ops->get_cpuid,
such as vendor_intel, and those should also use check_limit==true and
check the return value of ctxt->ops->get_cpuid.

Paolo