Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262545AbUCaVca (ORCPT ); Wed, 31 Mar 2004 16:32:30 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262585AbUCaVca (ORCPT ); Wed, 31 Mar 2004 16:32:30 -0500 Received: from parcelfarce.linux.theplanet.co.uk ([195.92.249.252]:40908 "EHLO www.linux.org.uk") by vger.kernel.org with ESMTP id S262545AbUCaV1E (ORCPT ); Wed, 31 Mar 2004 16:27:04 -0500 Message-ID: <406B3799.5060203@pobox.com> Date: Wed, 31 Mar 2004 16:26:49 -0500 From: Jeff Garzik User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030703 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Chris Mason CC: "Stephen C. Tweedie" , Bartlomiej Zolnierkiewicz , Jens Axboe , Linux Kernel Subject: Re: [PATCH] barrier patch set References: <20040319153554.GC2933@suse.de> <200403201723.11906.bzolnier@elka.pw.edu.pl> <1079800362.11062.280.camel@watt.suse.com> <200403201805.26211.bzolnier@elka.pw.edu.pl> <1080662685.1978.25.camel@sisko.scot.redhat.com> <1080674384.3548.36.camel@watt.suse.com> <1080683417.1978.53.camel@sisko.scot.redhat.com> <4069F2FC.90003@pobox.com> <1080742105.1991.40.camel@sisko.scot.redhat.com> <1080742895.3547.139.camel@watt.suse.com> In-Reply-To: <1080742895.3547.139.camel@watt.suse.com> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1099 Lines: 41 Chris Mason wrote: > On Wed, 2004-03-31 at 09:08, Stephen C. Tweedie wrote: > >>Hi, >> >>On Tue, 2004-03-30 at 23:21, Jeff Garzik wrote: >> >> >>>For IDE, O_DIRECT and O_SYNC can use special "FUA" commands, which don't >>>return until the data is on the platter. >> >>fsync() is still really nasty, because that can require that we wait on >>IO that was submitted by the VM before we knew that there was a >>synchronous IO wait coming. > > > Yes, it gets ugly in a hurry. Jeff, look at the whole thread about the > O_DIRECT read vs buffered write races. I don't think we can use FUA for Yes, I'm aware of the thread... > fsync or O_SYNC without using it for every write. Why not for O_SYNC? Is some crazy userspace application flipping this bit on and off rapidly? > We might be able to get away with using it on O_DIRECT. Nod. Jeff - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/