From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Sat, 12 May 2001 20:01:17 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Sat, 12 May 2001 20:01:07 -0400 Received: from mailhost.tue.nl ([131.155.2.5]:3385 "EHLO mailhost.tue.nl") by vger.kernel.org with ESMTP id ; Sat, 12 May 2001 20:00:56 -0400 Message-ID: <20010513020053.B7635@win.tue.nl> Date: Sun, 13 May 2001 02:00:53 +0200 From: Guest section DW To: Alan Cox , viro@math.psu.edu (Alexander Viro) Cc: szabi@inf.elte.hu (BERECZ Szabolcs), linux-kernel@vger.kernel.org Subject: Re: mount /dev/hdb2 /usr; swapon /dev/hdb2 keeps flooding In-Reply-To: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.93i In-Reply-To: ; from Alan Cox on Sun, May 13, 2001 at 12:32:20AM +0100 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Sun, May 13, 2001 at 12:32:20AM +0100, Alan Cox wrote: > > > root@kama3:/home/szabi# cat /proc/mounts > > > /dev/hdb2 /usr ext2 rw 0 0 > > > root@kama3:/home/szabi# swapon /dev/hdb2 > > > > - Doctor, it hurts when I do it! > > - Don't do it, then. > > > > Just what behaviour had you expected? > > EBUSY would be somewhat nicer. I suppose the return will be EINVAL when there is no swap signature. But the mistake is doing set_blocksize too early. (Even twice. Why?) We should first check whether things are OK before starting to make changes. Andries