From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S265492AbUHDNbZ (ORCPT ); Wed, 4 Aug 2004 09:31:25 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S265487AbUHDNbZ (ORCPT ); Wed, 4 Aug 2004 09:31:25 -0400 Received: from mail.parknet.co.jp ([210.171.160.6]:9744 "EHLO mail.parknet.co.jp") by vger.kernel.org with ESMTP id S265429AbUHDNbX (ORCPT ); Wed, 4 Aug 2004 09:31:23 -0400 To: walt Cc: Linux Kernel Subject: Re: [2.6.8-rc2-bk] New read/write bug in FAT fs References: <4110CF29.8060401@myrealbox.com> From: OGAWA Hirofumi Date: Wed, 04 Aug 2004 22:31:18 +0900 In-Reply-To: <4110CF29.8060401@myrealbox.com> Message-ID: <87657zkp21.fsf@devron.myhome.or.jp> User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.3.50 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org walt writes: > One of the changesets posted by Linus on August 2 introduced > a bug in the FAT fs: > > Even when a fat32 fs is mounted read-write I now get error > messages claiming the fs is 'read-only' when I try to write > to it. > > The only change I can see which fits the timing is to inode.c > which was posted on August 2. Apologies if I am blaming > the wrong changeset. > > Could someone confirm this bug for me? This is intention. The default codepage/iocharset which is easy to cause a mistake and unclear was deleted from fatfs. So default is mounted as read only. You need to explicitly specify the "codepage" and "iocharset" options. -- OGAWA Hirofumi