From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1LrDaf-0007MD-JK for qemu-devel@nongnu.org; Tue, 07 Apr 2009 11:51:57 -0400 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1LrDaa-0007Jw-0F for qemu-devel@nongnu.org; Tue, 07 Apr 2009 11:51:56 -0400 Received: from [199.232.76.173] (port=57482 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1LrDaZ-0007Jl-TI for qemu-devel@nongnu.org; Tue, 07 Apr 2009 11:51:51 -0400 Received: from verein.lst.de ([213.95.11.210]:51065) by monty-python.gnu.org with esmtps (TLS-1.0:DHE_RSA_3DES_EDE_CBC_SHA1:24) (Exim 4.60) (envelope-from ) id 1LrDaZ-0000rn-Jj for qemu-devel@nongnu.org; Tue, 07 Apr 2009 11:51:51 -0400 Date: Tue, 7 Apr 2009 17:51:44 +0200 From: Christoph Hellwig Subject: Re: [Qemu-devel] [PATCH 05/10] xen: add block device backend driver. Message-ID: <20090407155144.GA10537@lst.de> References: <1238621982-18333-1-git-send-email-kraxel@redhat.com> <1238621982-18333-6-git-send-email-kraxel@redhat.com> <20090402170209.GA10089@lst.de> <49DB2948.6050701@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <49DB2948.6050701@redhat.com> Reply-To: qemu-devel@nongnu.org List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Cc: xen-devel@lists.xensource.com, Christoph Hellwig On Tue, Apr 07, 2009 at 12:22:00PM +0200, Gerd Hoffmann wrote: > Hi, > > >We already have bdrv_aio_readv/writev which currently linearize the > >buffer underneath. Hopefully Anthony will have commited the patch to > >implement the real one while I'm writing this, too :) > > Is there any specific reason to keep the nb_sectors argument for > bdrv_aio_readv, other than cut&paste from the non-vectored versions? > I think it is superfluous now, it can be trivially calculated using > iov->size. For the functions in block.c I want to keep them the same as they were before, and for the methods I want to keep the signature the same as the exported wrappers. All this is going to change pretty soon I hope, we'll need another argument to bdrv_aio_write for barrier writes, and I might switch to a completely byte based interface for the aio requests with flags in the block driver state for flagging the alignment restrictions.