From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:41907) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1UpMLj-0000HI-ID for qemu-devel@nongnu.org; Wed, 19 Jun 2013 13:39:16 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1UpMLi-0002uy-7B for qemu-devel@nongnu.org; Wed, 19 Jun 2013 13:39:15 -0400 Received: from mail-wg0-x236.google.com ([2a00:1450:400c:c00::236]:41307) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1UpMLh-0002uu-Q4 for qemu-devel@nongnu.org; Wed, 19 Jun 2013 13:39:14 -0400 Received: by mail-wg0-f54.google.com with SMTP id n11so4692502wgh.21 for ; Wed, 19 Jun 2013 10:39:13 -0700 (PDT) Sender: Paolo Bonzini Message-ID: <51C1ECB9.6060201@redhat.com> Date: Wed, 19 Jun 2013 19:39:05 +0200 From: Paolo Bonzini MIME-Version: 1.0 References: <1371542991-15911-1-git-send-email-gaowanlong@cn.fujitsu.com> <1371542991-15911-4-git-send-email-gaowanlong@cn.fujitsu.com> <51C02665.1060205@redhat.com> <20130618190049.GA2825@otherpad.lan.raisama.net> <51C1653F.4000902@cn.fujitsu.com> In-Reply-To: <51C1653F.4000902@cn.fujitsu.com> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH 3/7] NUMA: parse guest numa nodes memory policy List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: gaowanlong@cn.fujitsu.com Cc: andre.przywara@amd.com, aliguori@us.ibm.com, Eduardo Habkost , qemu-devel@nongnu.org, Bandan Das , y-goto@jp.fujitsu.com, afaerber@suse.de Il 19/06/2013 10:01, Wanlong Gao ha scritto: > On 06/19/2013 04:19 AM, Bandan Das wrote: >> Eduardo Habkost writes: >> >>> On Tue, Jun 18, 2013 at 11:20:37AM +0200, Paolo Bonzini wrote: >>> [...] >>>> Also, please use QemuOpts instead of yet another homegrown parser. >>>> Eduardo, I think you had the most recent attempt to convert -numa to >>>> QemuOpts? >>> >>> I had one, but I believe it is more complex than it should have been. I >>> was creating a "numa-node" config section while keeping "-numa" just for >>> compatbility, but I don't think we really need to do that. >> >> Ah, I was working on an update to Eduardo's earlier proposals for multiple CPU ranges >> and part of the change was to convert to QemuOpts. >> >> Probably needs more testing but posted it anyway since we are already discussing this : >> [PATCH v3] vl.c: Support multiple CPU ranges on -numa option >> (hasn't shown up in the archives yet) > > Here is the archive: http://thread.gmane.org/gmane.comp.emulators.qemu/217491 > > So, are you all ACK with this? And we are not considering compatible thing by using > "cpu" instead of "cpus" here? No; as Eduardo pointed out, the "cpus" must be kept. But apart from that, picking up Bandan's patch in v2 of this series should be fine. Paolo