From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christoph Hellwig Subject: Re: adding proper O_SYNC/O_DSYNC, was Re: O_DIRECT and barriers Date: Fri, 28 Aug 2009 12:17:45 -0400 Message-ID: <20090828161745.GA8755@infradead.org> References: <20090821142635.GB30617@infradead.org> <20090821152459.GC6929@shareable.org> <20090821174525.GA28861@infradead.org> <20090822005006.GA22530@shareable.org> <20090824023422.GA775@infradead.org> <20090827143459.GB31453@shareable.org> <20090827171044.GA5427@infradead.org> <4A96C14C.8040105@redhat.com> <20090828154647.GA15808@infradead.org> <4A98008B.6050503@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Christoph Hellwig , Jamie Lokier , linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org To: Ulrich Drepper Return-path: Content-Disposition: inline In-Reply-To: <4A98008B.6050503@redhat.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org On Fri, Aug 28, 2009 at 09:06:35AM -0700, Ulrich Drepper wrote: > On 08/28/2009 08:46 AM, Christoph Hellwig wrote: >> - given that our current O_SYNC really is and always has been actuall >> Posix O_DSYNC > > If this is true, then this proposal would work, yes. I'll put it on my todo list. While reading through the Posix specs I came up with some questions that you might be able to answer: - O_RSYNC basically means we need to commit atime updates before a read returns, right? It would be easy to implement it in a slightly suboptimal fashion, but is there any point?