Message-ID: <58AD0094.90304@windriver.com>
Date: Tue, 21 Feb 2017 21:08:04 -0600
From: Chris Friesen <chris.friesen@windriver.com>
MIME-Version: 1.0
References: <589C7E96.9060905@huawei.com> <589D83CE.1090803@huawei.com>	
	<589DDC05.9010807@windriver.com> <58AA51D6.6020508@huawei.com>
	<1487565495.3740.27.camel@intel.com>
In-Reply-To: <1487565495.3740.27.camel@intel.com>
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Transfer-Encoding: 7bit
Subject: Re: [Qemu-devel] kvm bug in __rmap_clear_dirty during live migration
To: "Han, Huaitong" <huaitong.han@intel.com>, "hangaohuai@huawei.com" <hangaohuai@huawei.com>
Cc: "kvm@vger.kernel.org" <kvm@vger.kernel.org>, "fangying1@huawei.com" <fangying1@huawei.com>, "herongguang.he@huawei.com" <herongguang.he@huawei.com>, "xudong.hao@linux.intel.com" <xudong.hao@linux.intel.com>, "qemu-devel@nongnu.org" <qemu-devel@nongnu.org>, "pbonzini@redhat.com" <pbonzini@redhat.com>, "wangxinxin.wang@huawei.com" <wangxinxin.wang@huawei.com>, "kai.huang@linux.intel.com" <kai.huang@linux.intel.com>, "rkrcmar@redhat.com" <rkrcmar@redhat.com>, "guangrong.xiao@linux.intel.com" <guangrong.xiao@linux.intel.com>

On 02/19/2017 10:38 PM, Han, Huaitong wrote:
> Hi, Gaohuai
>
> I tried to debug the problem, and I found the indirect cause may be that
> the rmap value is not cleared when KVM mmu page is freed. I have read
> code without the root cause. Can you stable reproduce the the issue?
> Many guesses need to be verified.

In both cases it seems to have been triggered by repeatedly live-migrating a KVM 
virtual machine between two hypervisors with Broadwell CPUs running the latest 
CentOS 7.

It's a race of some sort, it doesn't happen every time.

Chris