From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:45046) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Vr5gg-0001yr-Qg for qemu-devel@nongnu.org; Thu, 12 Dec 2013 07:48:24 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Vr5ga-0007Xp-Vy for qemu-devel@nongnu.org; Thu, 12 Dec 2013 07:48:18 -0500 Received: from mx.ipv6.kamp.de ([2a02:248:0:51::16]:33243 helo=mx01.kamp.de) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Vr5ga-0007Xh-KZ for qemu-devel@nongnu.org; Thu, 12 Dec 2013 07:48:12 -0500 Message-ID: <52A9B087.4030804@kamp.de> Date: Thu, 12 Dec 2013 13:48:07 +0100 From: Peter Lieven MIME-Version: 1.0 References: <1386796109-15264-1-git-send-email-kwolf@redhat.com> <52A95324.2050003@kamp.de> <20131212094748.GF10007@dhcp-200-207.str.redhat.com> In-Reply-To: <20131212094748.GF10007@dhcp-200-207.str.redhat.com> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH 00/22] block: Support for 512b-on-4k emulation List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Kevin Wolf Cc: pbonzini@redhat.com, qemu-devel@nongnu.org, stefanha@redhat.com Am 12.12.2013 10:47, schrieb Kevin Wolf: > Am 12.12.2013 um 07:09 hat Peter Lieven geschrieben: >> Should it be possible to boot from a 4K-native drive with this series? >> If yes, I will run some test with some new iSCSI arrays we got for testing >> they can export 4k blocksize LUNs. > Yes, you should be able to use a 4k-native backend for a 512b-based > guest with this series (at the cost of some performance, of course). > > Thanks for your testing offer! > >> Anyway, can you please include the patch from Paolo which sets >> the bs->request_alignment = iscsilun->block_size in iscsi_open. >> It should be a one-liner. > Sure, I've included it now. For your test, it's probably easiest if you > pull the updated version from here: > > git://repo.or.cz/qemu/kevin.git align > > Do you think you'll run some tests already today? If so, I could wait a > bit with sending out a v2 and see what your first results are. Will take some more time. Found a but in the implementation of my storage which hit a bug in qemu. Will send a fix for the latter one shortly. Peter