From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757361AbYEOHjn (ORCPT ); Thu, 15 May 2008 03:39:43 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753421AbYEOHjb (ORCPT ); Thu, 15 May 2008 03:39:31 -0400 Received: from fxip-0047f.externet.hu ([88.209.222.127]:46185 "EHLO pomaz-ex.szeredi.hu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753042AbYEOHja (ORCPT ); Thu, 15 May 2008 03:39:30 -0400 To: viro@ZenIV.linux.org.uk CC: miklos@szeredi.hu, akpm@linux-foundation.org, hch@infradead.org, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org In-reply-to: <20080515072358.GS13907@ZenIV.linux.org.uk> (message from Al Viro on Thu, 15 May 2008 08:23:58 +0100) Subject: Re: [patch 08/24] vfs: immutable inode checking cleanup References: <20080506091327.259950960@szeredi.hu> <20080506091416.025653274@szeredi.hu> <20080515072358.GS13907@ZenIV.linux.org.uk> Message-Id: From: Miklos Szeredi Date: Thu, 15 May 2008 09:39:13 +0200 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org > > --- linux-2.6.orig/fs/attr.c 2008-05-06 11:04:29.000000000 +0200 > > +++ linux-2.6/fs/attr.c 2008-05-06 11:04:35.000000000 +0200 > > @@ -108,6 +108,12 @@ int notify_change(struct dentry * dentry > > struct timespec now; > > unsigned int ia_valid = attr->ia_valid; > > > > + if (ia_valid & (ATTR_MODE | ATTR_UID | ATTR_GID | > > + ATTR_ATIME_SET | ATTR_MTIME_SET)) { > > + if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) > > + return -EPERM; > > + } > > + > > now = current_fs_time(inode->i_sb); > > > > attr->ia_ctime = now; > > Index: linux-2.6/fs/utimes.c > > =================================================================== > > --- linux-2.6.orig/fs/utimes.c 2008-05-06 11:04:29.000000000 +0200 > > +++ linux-2.6/fs/utimes.c 2008-05-06 11:04:35.000000000 +0200 > > @@ -105,10 +105,6 @@ long do_utimes(int dfd, char __user *fil > > /* Don't worry, the checks are done in inode_change_ok() */ > > newattrs.ia_valid = ATTR_CTIME | ATTR_MTIME | ATTR_ATIME; > > if (times) { > > - error = -EPERM; > > - if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) > > - goto mnt_drop_write_and_out; > > - > > if (times[0].tv_nsec == UTIME_OMIT) > > newattrs.ia_valid &= ~ATTR_ATIME; > > else if (times[0].tv_nsec != UTIME_NOW) { > > Erm... What happens if we have UTIME_NOW on both and inode is marked > immutable? AFAICS, you've just switched from -EPERM to -EACCES. > For append-only it's even more interesting - you go from -EPERM to > success. Yup. These are defensible on the ground that times == NULL should be equivalent to UTIME_NOW on both. There's really no reason to be stricter on the latter case. > It might or might not be the right thing to do, and the syscall is > new, but... Do you have an ACK from Ulrich? No. I could CC him on these, but he's not even acking my security patches on utimesat, so I don't have high hopes. Miklos