From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Vrabel Subject: Re: Domain Save Image Format proposal (draft B) Date: Tue, 11 Feb 2014 17:10:42 +0000 Message-ID: <52FA5992.3070207@citrix.com> References: <52F90A71.40802@citrix.com> <21242.21650.571988.662930@mariner.uk.xensource.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <21242.21650.571988.662930@mariner.uk.xensource.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Ian Jackson Cc: Shriram Rajagopalan , Stefano Stabellini , Ian Campbell , "Xen-devel@lists.xen.org" List-Id: xen-devel@lists.xenproject.org On 11/02/14 16:49, Ian Jackson wrote: > David Vrabel writes ("Domain Save Image Format proposal (draft B)"): >> Here is a draft of a proposal for a new domain save image format. It >> does not currently cover all use cases (e.g., images for HVM guest are >> not considered). > > I think this is a good start. I've made some other comments already. > >> Overview >> ======== > > I would like to make another perhaps controversial suggestion. We > should explicitly specify that the receiver may advertise its > capabilities to the sender, so that backwards-migration _can_ be > supported if we choose to do so. > > In practice I think that means a capability advertisement block. > Probably, one bit per version, one bit per record type, etc. > > I greatly prefer doing forward-compatibility with new record type > enums etc. than with version numbers. Version numbers presuppose a > strict global order on all the implementations' capabilities, which is > of course not necessarily true in free software. I think I quite like this idea. There would still need to be a central repository of record types. David