Re: [PATCH v11 14/16] KVM: x86/vmx: Flip Arch LBREn bit on guest state change

From: Paolo Bonzini <pbonzini@redhat.com>
To: "Yang, Weijiang" <weijiang.yang@intel.com>,
	"jmattson@google.com" <jmattson@google.com>,
	"seanjc@google.com" <seanjc@google.com>,
	"kan.liang@linux.intel.com" <kan.liang@linux.intel.com>,
	"like.xu.linux@gmail.com" <like.xu.linux@gmail.com>,
	"vkuznets@redhat.com" <vkuznets@redhat.com>,
	"Wang, Wei W" <wei.w.wang@intel.com>,
	"kvm@vger.kernel.org" <kvm@vger.kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v11 14/16] KVM: x86/vmx: Flip Arch LBREn bit on guest state change
Date: Thu, 12 May 2022 15:18:40 +0200	[thread overview]
Message-ID: <8a15c4b4-cabe-7bc3-bd98-bd669d586616@redhat.com> (raw)
In-Reply-To: <9e2b5e9f-25a2-b724-c6d7-282dc987aa99@intel.com>

On 5/11/22 09:43, Yang, Weijiang wrote:
>>
>> Instead of using flip_arch_lbr_ctl, SMM should save the value of the MSR
>> in kvm_x86_ops->enter_smm, and restore it in kvm_x86_ops->leave_smm
>> (feel free to do it only if guest_cpuid_has(vcpu, X86_FEATURE_LM), i.e.
>> the 32-bit case can be ignored).
> 
> In the case of migration in SMM, I assume kvm_x86_ops->enter_smm() 
> called in source side
> 
> and kvm_x86_ops->leave_smm() is called at destination, then should the 
> saved LBREn be transferred
> 
> to destination too? The destination can rely on the bit to defer setting 
> LBREn bit in

Hi, it's transferred automatically if the MSR is saved in the SMM save 
state area.  Both enter_smm and leave_smm can access the save state area.

The enter_smm callback is called after saving "normal" state, and it has 
to save the state + clear the bit; likewise, the leave_smm callback is 
called before saving "normal" state and will restore the old value of 
the MSR.

Thanks,

Paolo

> VMCS until kvm_x86_ops->leave_smm() is called. is it good? thanks!