From: Keir Fraser Subject: Re: [rfc 00/18] ioemu: use devfn instead of slots as the unit for passthrough Date: Mon, 02 Mar 2009 10:08:29 +0000 Message-ID: References: <20090302095329.GA21933@verge.net.au> Mime-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20090302095329.GA21933@verge.net.au> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Simon Horman Cc: Yuji Shimada , "xen-devel@lists.xensource.com" , Ian Jackson List-Id: xen-devel@lists.xenproject.org On 02/03/2009 09:53, "Simon Horman" wrote: >> Not really what I had in mind. Xend can do the GSI->slot mapping, to ensure >> non-conflicting GSIs. I don't think any hypervisor changes are required, let >> alone substantial ones. > > Is the idea that xend would allocate a gsi to a device and > then pass that gsi along as part of the device configuration > to the device model? > > If so, I think something similar to what I wrote, but moved > into xend could work quite well. But I sense that wasn't what > you had in mind either. I mean that xend can pick a virtual devfn for the device that it knows has a non-conflicting GSI. This avoids any need for dynamic mapping between devfn and GSI (which would be more of a pain in the neck -- for example, your patch doesn't work because certain parts of BIOS info tables need to be dynamically generated, as currently they hardcode the devfn-GSI relationship). -- Keir