From: Jeremy Fitzhardinge Subject: Re: [RFC Patch] Support for making an E820 PCI hole in toolstack (xl + xm) Date: Mon, 15 Nov 2010 11:32:38 -0800 Message-ID: <4CE18AD6.5070102@goop.org> References: Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Keir Fraser Cc: stefano.stabellini@eu.citrix.com, xen-devel@lists.xensource.com, bruce.edge@gmail.com, gianni.tedesco@citrix.com, Konrad Rzeszutek Wilk List-Id: xen-devel@lists.xenproject.org On 11/15/2010 10:41 AM, Keir Fraser wrote: >> Oh, there was another reason that I so obvious that I completly forgot. DomU >> has no idea where the host PCI hole starts. In most cases it is at 3GB (or >> even >> further up - 3.5GB), but a quick look for 'Allocating PCI resources starting >> at' >> at Google shows that there are some that start at 1.2G. > Hm, true. We could give you access to XENMEM_machine_memory_map? It's not > really got any big secrets or privileged things in it. :-) > > Or, is there much disadvantage, to having a static really big PCI hole? Say > starting at 1GB? The advantage of this would be the ability to hotplug PCI > devices to a domU even across save/restore/migrate -- this may not work so > well if you commit yourself to the hole size of the original host, and the > restore/migrate target host has a bigger hole! Well, the other question is whether the devices have to have the same pfn as mfn within the hole. We're emulating the PCI config space anyway - couldn't we stick the passthrough PCI space at 3G regardless of where it is on the real hardware? J