From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1Ma588-0004Cz-RE for qemu-devel@nongnu.org; Sun, 09 Aug 2009 05:55:56 -0400 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1Ma582-0004Ay-Uz for qemu-devel@nongnu.org; Sun, 09 Aug 2009 05:55:55 -0400 Received: from [199.232.76.173] (port=46991 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1Ma582-0004Ai-IY for qemu-devel@nongnu.org; Sun, 09 Aug 2009 05:55:50 -0400 Received: from mx2.redhat.com ([66.187.237.31]:55277) by monty-python.gnu.org with esmtp (Exim 4.60) (envelope-from ) id 1Ma581-0004wX-Ft for qemu-devel@nongnu.org; Sun, 09 Aug 2009 05:55:49 -0400 Message-ID: <4A7E9E7F.2090901@redhat.com> Date: Sun, 09 Aug 2009 13:01:35 +0300 From: Avi Kivity MIME-Version: 1.0 References: <20090806163549.GA25594@lst.de> In-Reply-To: <20090806163549.GA25594@lst.de> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Subject: [Qemu-devel] Re: virtio-blk performance and MSI List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Christoph Hellwig , Anthony Liguori Cc: qemu-devel@nongnu.org, kvm@vger.kernel.org On 08/06/2009 07:35 PM, Christoph Hellwig wrote: > Michael suggested to me a while ago to try MSI with virtio-blk and I > played with this small patch: > > > Index: qemu-kvm/hw/virtio-blk.c > =================================================================== > --- qemu-kvm.orig/hw/virtio-blk.c > +++ qemu-kvm/hw/virtio-blk.c > @@ -416,6 +416,7 @@ VirtIODevice *virtio_blk_init(DeviceStat > s->vdev.get_config = virtio_blk_update_config; > s->vdev.get_features = virtio_blk_get_features; > s->vdev.reset = virtio_blk_reset; > + s->vdev.nvectors = 2; > s->bs = bs; > s->rq = NULL; > if (strlen(ps = (char *)drive_get_serial(bs))) > > which gave about 5% speedups on 4k sized reads and writes, see the full > iozone output I attached. Now getting the information about using > multiple MSI vectors from the command line to virtio-blk similar to how > virtio-net does seems extremly messy right now. Waiting for Gerd's > additional qdev patches to make it easier as a qdev property. > > Looks good. Anthony, I think this applies upstream? -- error compiling committee.c: too many arguments to function