From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:59048) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1UnuCA-0006Wl-8Y for qemu-devel@nongnu.org; Sat, 15 Jun 2013 13:23:23 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1UnuC7-0008CL-Bp for qemu-devel@nongnu.org; Sat, 15 Jun 2013 13:23:22 -0400 Received: from mail-ye0-f176.google.com ([209.85.213.176]:45872) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1UnuC7-0008CF-8O for qemu-devel@nongnu.org; Sat, 15 Jun 2013 13:23:19 -0400 Received: by mail-ye0-f176.google.com with SMTP id m14so485585yen.35 for ; Sat, 15 Jun 2013 10:23:18 -0700 (PDT) Sender: Paolo Bonzini Message-ID: <51BCA2FF.7090302@redhat.com> Date: Sat, 15 Jun 2013 13:23:11 -0400 From: Paolo Bonzini MIME-Version: 1.0 References: <1370404705-4620-1-git-send-email-gaowanlong@cn.fujitsu.com> <1370404705-4620-2-git-send-email-gaowanlong@cn.fujitsu.com> <20130605134505.GS2580@otherpad.lan.raisama.net> <51B6D025.3040606@cn.fujitsu.com> <20130611134017.GC2895@otherpad.lan.raisama.net> <51B922FE.8090109@cn.fujitsu.com> <20130613125019.GI2895@otherpad.lan.raisama.net> <51BA4891.6020108@redhat.com> <87mwqttsbl.fsf@codemonkey.ws> <51BA6EEA.20901@cn.fujitsu.com> In-Reply-To: <51BA6EEA.20901@cn.fujitsu.com> Content-Type: text/plain; charset=ISO-8859-15 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH 2/2] Add monitor command mem-nodes List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: gaowanlong@cn.fujitsu.com Cc: andre.przywara@amd.com, Anthony Liguori , Eduardo Habkost , qemu-devel@nongnu.org Il 13/06/2013 21:16, Wanlong Gao ha scritto: > On 06/14/2013 09:05 AM, Anthony Liguori wrote: >> Paolo Bonzini writes: >> >>> Il 13/06/2013 08:50, Eduardo Habkost ha scritto: >>>> I believe an interface based on guest physical memory addresses is more >>>> flexible (and even simpler!) than one that only allows binding of whole >>>> virtual NUMA nodes. >>> >>> And "-numa node" is already one, what about just adding "mem-path=/foo" >>> or "host_node=NN" suboptions? Then "-mem-path /foo" would be a shortcut >>> for "-numa node,mem-path=/foo". >>> >>> I even had patches to convert -numa to QemuOpts, I can dig them out if >>> your interested. >> >> Ack. This is a very reasonable thing to add. > > How about making "-numa node,membind=0" like options, and also provide a > QMP interface "numa_set guest_node_id mempolicy". So that we can set > the mempolicy not only for file backed memory but also for anon mapped > guest numa node. This is full numa support in QEMU as you said. I'm making > the patches now. Yup, that's exactly what I called "host_node". membind also makes sense, but make it mem-bind or mem-host-node for consistency with mem-path. Paolo