From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from verein.lst.de ([213.95.11.211]:35351 "EHLO newverein.lst.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751460AbcGAFdZ (ORCPT ); Fri, 1 Jul 2016 01:33:25 -0400 Date: Fri, 1 Jul 2016 07:32:27 +0200 From: Christoph Hellwig To: Damien Le Moal Cc: Christoph Hellwig , Amir Goldstein , Ming Lei , linux-fsdevel , linux-block , Jens Axboe Subject: Re: Question about loop device DIO/AIO and FLUSH Message-ID: <20160701053227.GA917@lst.de> References: <20160701042510.GA381@lst.de> <20160701050455.GA828@lst.de> <3D8F3502-8EA7-4568-882B-90E48C8F6BCF@hgst.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <3D8F3502-8EA7-4568-882B-90E48C8F6BCF@hgst.com> Sender: linux-block-owner@vger.kernel.org List-Id: linux-block@vger.kernel.org On Fri, Jul 01, 2016 at 05:17:28AM +0000, Damien Le Moal wrote: > Thank you for the clarification. Makes sense. > Just to be picky though, I think that "completed I/Os" applies only > to flush in this case (flush of the volatile device cache to media). > fsync of the host cache (page cache most of the time) to the device > will issue new BIOs, and those will be waited for until completed. > Right ? For fsync the completed I/O at the syscall / VFS level. Fsync only needs to cover read/write calls that have return to the user, be that through a synchronous call or aio completion.