Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Sat, 4 Nov 2000 15:49:50 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Sat, 4 Nov 2000 15:49:39 -0500 Received: from spock.linux.it ([151.99.137.27]:49400 "HELO spock.linux.it") by vger.kernel.org with SMTP id ; Sat, 4 Nov 2000 15:49:36 -0500 Date: Sat, 4 Nov 2000 19:49:37 +0100 From: "Marco d'Itri" To: "Stephen C. Tweedie" Cc: Kernel Mailing List , linux-fsdevel@vger.kernel.org Subject: Re: Linux-2.4.0-test10 Message-ID: <20001104194937.E3423@wonderland.linux.it> In-Reply-To: <20001102171717.L1876@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <20001102171717.L1876@redhat.com>; from sct@redhat.com on Thu, Nov 02, 2000 at 05:17:17PM +0000 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Nov 02, "Stephen C. Tweedie" wrote: >2.2 O_SYNC is actually broken too --- it doesn't sync all metadata (in >particular, it doesn't update the inode), but I'd rather fix that for >2.4 rather than change 2.2, as the main users of O_SYNC, databases, >are writing to preallocated files anyway. What about fsync(2)? Will it update metadata too? -- ciao, Marco - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/