From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([140.186.70.92]:58243) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1QbvUx-00076M-TN for Qemu-devel@nongnu.org; Wed, 29 Jun 2011 10:12:12 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1QbvUu-0000yh-SZ for Qemu-devel@nongnu.org; Wed, 29 Jun 2011 10:12:11 -0400 Received: from verein.lst.de ([213.95.11.211]:54619 helo=newverein.lst.de) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1QbvUu-0000yQ-E0 for Qemu-devel@nongnu.org; Wed, 29 Jun 2011 10:12:08 -0400 Date: Wed, 29 Jun 2011 16:12:06 +0200 From: Christoph Hellwig Message-ID: <20110629141206.GA16311@lst.de> References: <4E0B1399.4050807@redhat.com> <4E0B152B.1000201@codemonkey.ws> <4E0B1787.30709@redhat.com> <4E0B1943.4090400@codemonkey.ws> <4E0B32A9.3010809@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4E0B32A9.3010809@redhat.com> Subject: Re: [Qemu-devel] Default cache mode List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Kevin Wolf Cc: Stefan Hajnoczi , Qemu-devel@nongnu.org, Frediano Ziglio , Christoph Hellwig , Avi Kivity On Wed, Jun 29, 2011 at 04:11:53PM +0200, Kevin Wolf wrote: > If the guest is acting correctly, then this assumption is definitely > true with cache=none/writeback. What needs to happen is that the guest > issues a disk flush before sending your "transaction ok" message. If it > doesn't do that it can fail even on real hardware. cache=writethrough > makes things safe that aren't even safe on real hardware, that's why I > consider it extremely conservative. Depends on your defintion of real hardware. Traditionally disks always had WCE=0 semantics. When ATA disks added volatile writecaches and people started seeing problems due to it the advertisement of cache modes and cache flush commands were added. For non-ATA disks like SAS or FC WCE=1 still is the typical delivery setting.