From: "Duane Griffin" Subject: Re: [PATCH 4/6] ext2: tighten restrictions on inode flags Date: Thu, 12 Jun 2008 10:38:53 +0100 Message-ID: References: <1213142649-32685-1-git-send-email-duaneg@dghda.com> <1213142649-32685-2-git-send-email-duaneg@dghda.com> <1213142649-32685-3-git-send-email-duaneg@dghda.com> <1213142649-32685-4-git-send-email-duaneg@dghda.com> <1213142649-32685-5-git-send-email-duaneg@dghda.com> <20080612090629.GX3726@webber.adilger.int> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: akpm@linux-foundation.org, linux-ext4@vger.kernel.org, linux-kernel@vger.kernel.org To: "Andreas Dilger" Return-path: Received: from wf-out-1314.google.com ([209.85.200.174]:7210 "EHLO wf-out-1314.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752863AbYFLJix (ORCPT ); Thu, 12 Jun 2008 05:38:53 -0400 Received: by wf-out-1314.google.com with SMTP id 27so3533654wfd.4 for ; Thu, 12 Jun 2008 02:38:53 -0700 (PDT) In-Reply-To: <20080612090629.GX3726@webber.adilger.int> Content-Disposition: inline Sender: linux-ext4-owner@vger.kernel.org List-ID: 2008/6/12 Andreas Dilger : > On Jun 11, 2008 01:04 +0100, Duane Griffin wrote: >> At the moment there are few restrictions on which flags may be set on which >> inodes. Specifically DIRSYNC may only be set on directories and IMMUTABLE >> and APPEND may not be set on links. Tighten that to disallow TOPDIR being >> set on non-directories and SECRM, UNRM, COMPR, SYNC, DIRTY, COMPRBLK, >> NOCOMP, ECOMPR, INDEX, JOURNAL_DATA and NOTAIL being set on anything but >> regular files or directories. >> >> Introduce a flags masking function which masks flags based on mode and use >> it during inode creation and when flags are set via the ioctl to facilitate >> future consistency. > > This second set of patches is missing out on the presence of the > "EXT2_FL_USER_MODIFIABLE" mask in ext2_ioctl(EXT2_IOC_SETFLAGS). This is > what prevents "unsettable" flags from being set from userspace. > > I don't have any objection to additional filtering to avoid setting the > USER_MODIFIABLE flags on special files. OK, thanks. I convinced myself that the patch wouldn't actually change behaviour in the ioctl case, but thought it best to use it anyway for consistency and to avoid future problems. I should probably have mentioned that in the ChangeLog; I never know quite how exhaustive to be in these things... > Cheers, Andreas Cheers, Duane. -- "I never could learn to drink that blood and call it wine" - Bob Dylan