From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dario Faggioli Subject: Re: NUMA TODO-list for xen-devel Date: Wed, 08 Aug 2012 09:43:28 +0200 Message-ID: <1344411808.5763.6.camel@Solace> References: <1343837796.4958.32.camel@Solace> <501BA1C0.7040100@amd.com> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============5389100668122147790==" Return-path: In-Reply-To: <501BA1C0.7040100@amd.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: Andre Przywara Cc: Anil Madhavapeddy , George Dunlap , Andrew Cooper , xen-devel , Jan Beulich , "Zhang, Yang Z" List-Id: xen-devel@lists.xenproject.org --===============5389100668122147790== Content-Type: multipart/signed; micalg="pgp-sha1"; protocol="application/pgp-signature"; boundary="=-0wsv0xJ54i/q+cWLa42t" --=-0wsv0xJ54i/q+cWLa42t Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Fri, 2012-08-03 at 12:02 +0200, Andre Przywara wrote: > On 08/01/2012 06:16 PM, Dario Faggioli wrote: > > ... > > > > * automatic placement of Dom0, if possible (my current series i= s > > only affecting DomU) >=20 > I think Dom0 NUMA awareness should be one of the top priorities. If I=20 > boot my 8-node box with Xen, I end up with a NUMA-clueless Dom0 which=20 > actually has memory from all 8 nodes and thinks it's memory is flat. > Ok, I updated the Wiki page with a link to this (sub)thread --- more specifically, the mails where we agree about the new syntax. I can work on it, but not in the next few days, so let's see if anyone steps up before I get to look at it. :-) > > [D] - NUMA aware scheduling in Xen. Don't pin vcpus on nodes' pcpus, > > just have them _prefer_ running on the nodes where their memory > > is. >=20 > This would be really cool. I once thought about something like a=20 > home-node. We start with placement to allocate memory from one node.=20 > Then we relax the VCPU-pinning, but mark this node as special for this= =20 > guest, so that it if possible happens to get run there. But in times of= =20 > CPU pressure we are happy to let it run on other nodes: CPU starving is= =20 > much worse than NUMA penalty. >=20 Yep. Patches coming soon. Thanks and Regards, Dario --=20 <> (Raistlin Majere) ----------------------------------------------------------------- Dario Faggioli, Ph.D, http://retis.sssup.it/people/faggioli Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK) --=-0wsv0xJ54i/q+cWLa42t Content-Type: application/pgp-signature; name="signature.asc" Content-Description: This is a digitally signed message part Content-Transfer-Encoding: 7bit -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (GNU/Linux) iEYEABECAAYFAlAiGKAACgkQk4XaBE3IOsSoEwCeNpwhfSJ/P/Xp+3W0CbU7dIWT w5QAniTYfdWGC2XmztdNrVli58PzRHvm =aNqW -----END PGP SIGNATURE----- --=-0wsv0xJ54i/q+cWLa42t-- --===============5389100668122147790== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel --===============5389100668122147790==--