From mboxrd@z Thu Jan 1 00:00:00 1970 From: Anthony Liguori Subject: Re: [PATCH] XML-RPC: Cope with large integers on x86-64 systems Date: Tue, 28 Mar 2006 09:18:24 -0600 Message-ID: <442953C0.4060200@us.ibm.com> References: <44289F79.9000501@us.ibm.com> <9bf498105410d761383ca3185d5f7369@cl.cam.ac.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <9bf498105410d761383ca3185d5f7369@cl.cam.ac.uk> 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: xen-devel , Ewan Mellor List-Id: xen-devel@lists.xenproject.org Keir Fraser wrote: > > On 28 Mar 2006, at 03:29, Anthony Liguori wrote: > >> Attached patch fixes a problem in marshaling large integers. It >> seems to only occur on systems with very large memory. However, on >> those systems, it's a show-stopper. I see this as a temporary >> solution until we can sanitize the XML-RPC functions and get rid of >> all the S-Expression stuff. > > I thought the main reason to push the XML-RPC stuff before 3.0.2 was > to give a measure of stability to that interface, and provide a more > supportable interface for vendors to lock onto? That reasoning doesn't > really hold if the core of the interface needs more refactoring and > cleaning up post 3.0.2. The approach for the XML-RPC work was to first get XML-RPC support into the tree and then build a standard interface. Building a standard interface requires refactoring (which will be simplification) to get rid of things like S-Expression configs. To get rid of that, we have to get rid of the S-Expression/HTTP layer. This whole reasoning is outlined on the wiki page. http://wiki.xensource.com/xenwiki/Xend/XML-RPC Regards, Anthony Liguori > -- Keir >