From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:51106) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dNlPd-00013B-79 for qemu-devel@nongnu.org; Wed, 21 Jun 2017 15:35:38 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dNlPa-0003X9-4T for qemu-devel@nongnu.org; Wed, 21 Jun 2017 15:35:37 -0400 Received: from mail5.windriver.com ([192.103.53.11]:34566 helo=mail5.wrs.com) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1dNlPZ-0003WS-RK for qemu-devel@nongnu.org; Wed, 21 Jun 2017 15:35:34 -0400 Received: from ALA-HCA.corp.ad.wrs.com (ala-hca.corp.ad.wrs.com [147.11.189.40]) by mail5.wrs.com (8.15.2/8.15.2) with ESMTPS id v5LJZUwD032074 (version=TLSv1 cipher=AES128-SHA bits=128 verify=OK) for ; Wed, 21 Jun 2017 12:35:30 -0700 Message-ID: <594ACA81.3010402@windriver.com> Date: Wed, 21 Jun 2017 13:35:29 -0600 From: Chris Friesen MIME-Version: 1.0 References: <20170621191601.GB2425@work-vm> In-Reply-To: <20170621191601.GB2425@work-vm> Content-Type: text/plain; charset="utf-8"; format=flowed Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [dpdk-dev] Will huge page have negative effect on guest vm in qemu enviroment? List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org On 06/21/2017 01:16 PM, Dr. David Alan Gilbert wrote: > * Sam (batmanustc@gmail.com) wrote: >> Thank you~ >> >> 1. We have a compare test on qemu-kvm enviroment with huge page and wi= thout >> huge page. Qemu start process is much longer in huge page enviromwnt. = And I >> write an email titled with '[DPDK-memory] how qemu waste such long tim= e >> under dpdk huge page envriment?'. I could resend it later. > >> 2. Then I have another test on qemu-kvm enviroment with huge page and >> without huge page, which I didn't start ovs-dpdk and vhostuser port in= qemu >> start process. And I found Qemu start process is also much longer in h= uge >> page enviroment. >> >> So I think huge page enviroment, which grub2.cfg file is specified in >> =E2=80=98[DPDK-memory] >> how qemu waste such long time under dpdk huge page envriment?=E2=80=99= , will really >> have negative effect on qemu start up process. >> >> That's why we don't like to use ovs-dpdk. Althrough ovs-dpdk is faster= , but >> the start up process of qemu is much longer then normal ovs, and the r= eason >> is nothing with ovs but huge page. For customers, vm start up time is >> important then network speed. > > How are you setting up hugepages? What values are you putting in the > various /proc or cmdline options and how are you specifying them on > QEMU's commandline. > > I think one problem is that with hugepages qemu normally allocates them > all at the start; I think there are cases where that means moving a lo= t > of memory about, especially if you lock it to particular NUMA nodes. For what it's worth, we use something like this: -object=20 memory-backend-file,id=3Dram-node0,prealloc=3Dyes,mem-path=3D/mnt/huge-20= 48kB/libvirt/qemu,share=3Dyes,size=3D2147483648,host-nodes=3D0,policy=3Db= ind=20 -numa node,nodeid=3D0,cpus=3D0-3,memdev=3Dram-node0 and I haven't noticed it taking particularly long to start up. All our=20 hugepages are reserved at host startup though. Chris