From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ian Campbell Subject: Re: (v2) Design proposal for RMRR fix Date: Mon, 12 Jan 2015 13:57:35 +0000 Message-ID: <1421071055.26317.72.camel@citrix.com> References: <1421070128.26317.61.camel@citrix.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" 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: "Tian, Kevin" Cc: "wei.liu2@citrix.com" , "stefano.stabellini@eu.citrix.com" , "george.dunlap@eu.citrix.com" , "tim@xen.org" , "ian.jackson@eu.citrix.com" , "xen-devel@lists.xen.org" , Jan Beulich , "Zhang, Yang Z" , "Chen, Tiejun" List-Id: xen-devel@lists.xenproject.org On Mon, 2015-01-12 at 13:53 +0000, Tian, Kevin wrote: > > From: Ian Campbell [mailto:Ian.Campbell@citrix.com] > > Sent: Monday, January 12, 2015 9:42 PM > > > > On Fri, 2015-01-09 at 06:57 +0000, Tian, Kevin wrote: > > > 3) report-sel vs. report-all > > > > One thing I'm not clear on is whether you are suggesting to reserve RMRR > > (either -all or -sel) for every domain by default, or whether the guest > > CFG will need to explicitly opt-in, IOW is there a 3rd report-none > > option which is the default unless otherwise requested (e.g. by > > e820_host=1, or some other new option)? > > only when a device is assigned (or potentially prepare for hotplug usage), How is this triggered though? Via pci= non-empty in the guest CFG? This sounds like it should behave similarly (in terms of when it is enabled, not necessarily in functionality) to the e820_host boolean, which is set by default if the pci= list is not empty, or can be overridden if the host admin anticipates doing hotplug. Ian.