From mboxrd@z Thu Jan 1 00:00:00 1970 From: George Dunlap Subject: Re: [PATCH V4] X86/vMCE: handle broken page with regard to migration Date: Mon, 3 Dec 2012 11:24:07 +0000 Message-ID: <50BC8BD7.3020908@eu.citrix.com> References: <1354183356.25834.108.camel@zakaz.uk.xensource.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; Format="flowed" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: "Liu, Jinsong" Cc: "xen-devel@lists.xensource.com" , Ian Jackson , Ian Campbell , Jan Beulich List-Id: xen-devel@lists.xenproject.org On 30/11/12 18:51, Liu, Jinsong wrote: > > Not exactly I think. Reading the broken page will trigger a serious SRAR error. Under such case hypervisor will inject a vMCE to the guest which was migrating, not dom0. The reason of this injection is, guest is best one to handle it, w/ sufficient clue/status/information (other component like hypervisor/dom0 are not proper). For xl migration process, after return from MCE context, it *again* read the broken page ... this will kill system entirely --> so we definitely not care migration any more. Why would the vMCE be sent to the guest, rather than the vcpu that was running when the SRAR error occured (i.e., the vcpu on which the migration process was running)? -George