From mboxrd@z Thu Jan 1 00:00:00 1970 From: Anthony Liguori Subject: Re: [Qemu-devel] Ideas wiki for GSoC 2010 Date: Mon, 15 Mar 2010 08:23:45 -0500 Message-ID: <4B9E34E1.3090709@codemonkey.ws> References: <20100310183023.6632aece@redhat.com> <4B9E2745.7060903@redhat.com> <20100315125313.GK9457@il.ibm.com> <20100315130310.GE13108@8bytes.org> <4B9E320E.7040605@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Joerg Roedel , Muli Ben-Yehuda , Luiz Capitulino , qemu-devel@nongnu.org, aliguori@us.ibm.com, kvm@vger.kernel.org, jan.kiszka@siemens.com, agraf@suse.de, agl@us.ibm.com, Nadav Amit , Ben-Ami Yassour1 To: Avi Kivity Return-path: Received: from mail-pv0-f174.google.com ([74.125.83.174]:37404 "EHLO mail-pv0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S964876Ab0CONXt (ORCPT ); Mon, 15 Mar 2010 09:23:49 -0400 Received: by pvg7 with SMTP id 7so893993pvg.19 for ; Mon, 15 Mar 2010 06:23:47 -0700 (PDT) In-Reply-To: <4B9E320E.7040605@redhat.com> Sender: kvm-owner@vger.kernel.org List-ID: On 03/15/2010 08:11 AM, Avi Kivity wrote: > On 03/15/2010 03:03 PM, Joerg Roedel wrote: >> >>>> I will add another project - iommu emulation. Could be very useful >>>> for doing device assignment to nested guests, which could make >>>> testing a lot easier. >>> Our experiments show that nested device assignment is pretty much >>> required for I/O performance in nested scenarios. >> Really? I did a small test with virtio-blk in a nested guest (disk read >> with dd, so not a real benchmark) and got a reasonable read-performance >> of around 25MB/s from the disk in the l2-guest. >> > > Your guest wasn't doing a zillion VMREADs and VMWRITEs every exit. > > I plan to reduce VMREAD/VMWRITE overhead for kvm, but not much we can > do for other guests. VMREAD/VMWRITEs are generally optimized by hypervisors as they tend to be costly. KVM is a bit unusual in terms of how many times the instructions are executed per exit. Regards, Anthony Liguori From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1NrAGt-0007co-JZ for qemu-devel@nongnu.org; Mon, 15 Mar 2010 09:23:51 -0400 Received: from [199.232.76.173] (port=44410 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1NrAGt-0007cM-75 for qemu-devel@nongnu.org; Mon, 15 Mar 2010 09:23:51 -0400 Received: from Debian-exim by monty-python.gnu.org with spam-scanned (Exim 4.60) (envelope-from ) id 1NrAGr-0007lF-Pq for qemu-devel@nongnu.org; Mon, 15 Mar 2010 09:23:51 -0400 Received: from mail-pw0-f45.google.com ([209.85.160.45]:57304) by monty-python.gnu.org with esmtp (Exim 4.60) (envelope-from ) id 1NrAGr-0007lB-G4 for qemu-devel@nongnu.org; Mon, 15 Mar 2010 09:23:49 -0400 Received: by pwi9 with SMTP id 9so1765490pwi.4 for ; Mon, 15 Mar 2010 06:23:47 -0700 (PDT) Message-ID: <4B9E34E1.3090709@codemonkey.ws> Date: Mon, 15 Mar 2010 08:23:45 -0500 From: Anthony Liguori MIME-Version: 1.0 Subject: Re: [Qemu-devel] Ideas wiki for GSoC 2010 References: <20100310183023.6632aece@redhat.com> <4B9E2745.7060903@redhat.com> <20100315125313.GK9457@il.ibm.com> <20100315130310.GE13108@8bytes.org> <4B9E320E.7040605@redhat.com> In-Reply-To: <4B9E320E.7040605@redhat.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Avi Kivity Cc: Muli Ben-Yehuda , agraf@suse.de, aliguori@us.ibm.com, kvm@vger.kernel.org, jan.kiszka@siemens.com, Joerg Roedel , qemu-devel@nongnu.org, Luiz Capitulino , agl@us.ibm.com, Nadav Amit , Ben-Ami Yassour1 On 03/15/2010 08:11 AM, Avi Kivity wrote: > On 03/15/2010 03:03 PM, Joerg Roedel wrote: >> >>>> I will add another project - iommu emulation. Could be very useful >>>> for doing device assignment to nested guests, which could make >>>> testing a lot easier. >>> Our experiments show that nested device assignment is pretty much >>> required for I/O performance in nested scenarios. >> Really? I did a small test with virtio-blk in a nested guest (disk read >> with dd, so not a real benchmark) and got a reasonable read-performance >> of around 25MB/s from the disk in the l2-guest. >> > > Your guest wasn't doing a zillion VMREADs and VMWRITEs every exit. > > I plan to reduce VMREAD/VMWRITE overhead for kvm, but not much we can > do for other guests. VMREAD/VMWRITEs are generally optimized by hypervisors as they tend to be costly. KVM is a bit unusual in terms of how many times the instructions are executed per exit. Regards, Anthony Liguori