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;
Message-ID: <5BA0CC1E.6030202@intel.com>
Date:   Tue, 18 Sep 2018 17:57:50 +0800
From:   Wei Wang <wei.w.wang@intel.com>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.7.0
MIME-Version: 1.0
To:     Andi Kleen <ak@linux.intel.com>,
        "Gonglei (Arei)" <arei.gonglei@huawei.com>
CC:     "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "kvm@vger.kernel.org" <kvm@vger.kernel.org>,
        "pbonzini@redhat.com" <pbonzini@redhat.com>,
        "kan.liang@intel.com" <kan.liang@intel.com>,
        "peterz@infradead.org" <peterz@infradead.org>,
        "mingo@redhat.com" <mingo@redhat.com>,
        "rkrcmar@redhat.com" <rkrcmar@redhat.com>,
        "like.xu@intel.com" <like.xu@intel.com>
Subject: Re: [PATCH v2 7/8] KVM: PMU: support to save/restore the guest lbr
 stack on vCPU switching
References: <1536233456-12173-1-git-send-email-wei.w.wang@intel.com> <1536233456-12173-8-git-send-email-wei.w.wang@intel.com> <33183CC9F5247A488A2544077AF19020DB0F1D5F@dggeml511-mbx.china.huawei.com> <20180918025616.GA10360@tassilo.jf.intel.com>
In-Reply-To: <20180918025616.GA10360@tassilo.jf.intel.com>
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 09/18/2018 10:56 AM, Andi Kleen wrote:
>>> From: Like Xu <like.xu@intel.com>
>>>
>>> This patch adds support to KVM to save/restore the lbr stack on vCPU
>>> context switching.
>>>
>>> When the guest sets the ACTIVE bit of MSR_KVM_PV_LBR_CTRL, a perf event
>>> is created on the host for the related vCPU. This perf event ensures the
>>> LBR stack to be saved/restored when the vCPU thread is scheduled out/in.
>>> The perf event is removed and freed when the guest clears the ACTIVE
>>> bit.
>>>
>> What about live migration? Does LBR stack need to be saved on the source side and
>> restored on the dest side with the passthrough mode?
> Yes it should. Either for call stack LBR, or when it is frozen/disabled.
>
> When it's not frozen/disabled and not in call stack LBR mode it likely doesn't
> hurt either, but it's not strictly needed because it will
> be replaced so quickly.

Yes, should be supported. We are working on v3 with the suggested lazy 
save approach, and will send it out shortly.

Best,
Wei