From mboxrd@z Thu Jan 1 00:00:00 1970 From: Paolo Bonzini Subject: Re: virtIO disks not usable Date: Mon, 03 Sep 2012 08:41:31 +0200 Message-ID: <5044511B.6090809@redhat.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: kvm@vger.kernel.org To: Andrew Holway Return-path: Received: from mail-pz0-f46.google.com ([209.85.210.46]:54639 "EHLO mail-pz0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751805Ab2ICGlf (ORCPT ); Mon, 3 Sep 2012 02:41:35 -0400 Received: by dady13 with SMTP id y13so3477847dad.19 for ; Sun, 02 Sep 2012 23:41:35 -0700 (PDT) In-Reply-To: Sender: kvm-owner@vger.kernel.org List-ID: Il 31/08/2012 15:19, Andrew Holway ha scritto: > I have 4096 block devices from sda..sdx. Do you mean (here and in your other cache=none message) that the host devices have 4k logical blocks? If so, and if you use cache=none, you need to export them as 4k block devices to the guest as well. This is done with the "logical_block_size=4096" sub-option in the "-device virtio-blk-pci" command-line option. Libvirt however does not support this yet. Alternatively, there were QEMU-side patches to support emulating a 512-over-4k device (using read-modify-write). One version I posted is at http://lists.gnu.org/archive/html/qemu-devel/2011-12/msg01570.html -- I need to rebase and repost. Paolo