From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from [140.186.70.92] (port=55393 helo=eggs.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1PqPLh-0004Tc-5q for qemu-devel@nongnu.org; Fri, 18 Feb 2011 07:22:14 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1PqPLg-0002VF-18 for qemu-devel@nongnu.org; Fri, 18 Feb 2011 07:22:13 -0500 Received: from mx1.redhat.com ([209.132.183.28]:5263) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1PqPLf-0002Up-Qm for qemu-devel@nongnu.org; Fri, 18 Feb 2011 07:22:11 -0500 Message-ID: <4D5E64D7.4090200@redhat.com> Date: Fri, 18 Feb 2011 13:23:51 +0100 From: Kevin Wolf MIME-Version: 1.0 Subject: Re: [Qemu-devel] [PATCH 3/3 v2] block/nbd: Make the NBD block device use the AIO interface References: <1297712422.12551.2.camel@den> <4D5A5ECD.7060701@redhat.com> <1297805193.12551.39.camel@den> <4D5BBC7B.9020807@redhat.com> <1297960482.21300.62.camel@desk4.office.bytemark.co.uk> <1297970917.12551.56.camel@den> <1298031391.17006.87.camel@desk4.office.bytemark.co.uk> In-Reply-To: <1298031391.17006.87.camel@desk4.office.bytemark.co.uk> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Nicholas Thomas Cc: stefanha@gmail.com, qemu-devel@nongnu.org Am 18.02.2011 13:16, schrieb Nicholas Thomas: > On Thu, 2011-02-17 at 19:28 +0000, Nicholas Thomas wrote: > >> Additional testing has revealed that this code breaks the stock >> nbd-server (the one on sourceforge) when large (well, 1.3MiB) write >> requests are sent to it. > > ....NBD has a limit of 1MB on the size of write requests. > NBD_BUFFER_SIZE in qemu-nbd.c - and I'm sure that's what's knocking out > the standard NBD server too. > > I didn't see any option to tell QEMU to split up writes to a certain > size before handing them off to the block driver, so I split the writes > up into multiple acbs. Reworked patch: Splitting it up manually sounds right. I haven't had a close look at your patches yet, but one thing that I noticed is that your patches are corrupted by line wraps. Please consider using git-send-email to avoid this kind of trouble or configure your mailer so that it stops doing this. Kevin