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: <4dc14170-8e22-9da7-30ec-a5597acf1f8e@gmail.com>
Date:   Fri, 26 May 2023 09:49:41 +0800
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:102.0) Gecko/20100101
 Thunderbird/102.10.0
Subject: Re: [PATCH v2 5/6] KVM: x86: Keep a per-VM MTRR state
Content-Language: en-US
To:     Sean Christopherson <seanjc@google.com>
Cc:     Yan Zhao <yan.y.zhao@intel.com>, kvm@vger.kernel.org,
        linux-kernel@vger.kernel.org, pbonzini@redhat.com
References: <20230509134825.1523-1-yan.y.zhao@intel.com>
 <20230509135300.1855-1-yan.y.zhao@intel.com>
 <3f09e751-33fd-7d60-78cd-6857d113e8bd@gmail.com>
 <ZGxbat2mM6AfOOVv@yzhao56-desk.sh.intel.com>
 <393b16f7-8359-5d77-7d5d-8942de987331@gmail.com>
 <ZG94Kmb8jMZKhtJW@google.com>
From:   Robert Hoo <robert.hoo.linux@gmail.com>
In-Reply-To: <ZG94Kmb8jMZKhtJW@google.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk

On 5/25/2023 11:00 PM, Sean Christopherson wrote:
[...]
> The MTRRs are not system wide or per-package though, they are per logical CPU.
> Yes, they "need" to be consistent with respect to one another, but only when the
> CPU is actually accessing memory.  This is a big reason why trying to track MTRRs
> as a per-VM asset in KVM is so difficult/messy.  Software doesn't rendezvous all
> CPUs and then do the write on just one CPU, each CPU does its own writes more or
> less independently.

Ah, got it, thanks!

(Some things of each logical processor seems just a shadow of an 
consolidate global one, e.g. CR0.CD. Some things are really separated 
setting of each logical processor, e.g. MTRR above. Really unfathomable ????)
> 
>> BTW, with regard to KVM_X86_QUIRK_CD_NW_CLEARED, I see svm honors it while
>> vmx doesn't before it clear CR0.CD/NW.
>>
>> svm_set_cr0():
>>
>> 	if (kvm_check_has_quirk(vcpu->kvm, KVM_X86_QUIRK_CD_NW_CLEARED))
>> 		hcr0 &= ~(X86_CR0_CD | X86_CR0_NW);
>>
>>
>> vmx_set_cr0():
>>
>> 	hw_cr0 = (cr0 & ~KVM_VM_CR0_ALWAYS_OFF);
>>
>> Perhaps vmx side can be fixed passingly?
> 
> Sadly, no.  SVM and VMX manage guest memtype completely differently.  VMX doesn't
> allow CR0.CD=1 when VMX is enabled, and so KVM needs to emulate CR0.CD via the EPT
> memtype.

OK, get it, thanks. Wasn't aware of this through SDM.