[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20170717164554.GD22195@quack2.suse.cz>
Date: Mon, 17 Jul 2017 18:45:54 +0200
From: Jan Kara <jack@...e.cz>
To: Ernesto A. Fernández
<ernesto.mnd.fernandez@...il.com>
Cc: Jan Kara <jack@...e.com>, Theodore Ts'o <tytso@....edu>,
Andreas Dilger <adilger.kernel@...ger.ca>,
Dave Kleikamp <shaggy@...nel.org>, linux-ext4@...r.kernel.org,
jfs-discussion@...ts.sourceforge.net,
reiserfs-devel@...r.kernel.org
Subject: Re: [PATCH 5/5] reiserfs: preserve i_mode if __reiserfs_set_acl()
fails
On Wed 12-07-17 06:56:04, Ernesto A. Fernández wrote:
> When changing a file's acl mask, __reiserfs_set_acl() will first set the
> group bits of i_mode to the value of the mask, and only then set the
> actual extended attribute representing the new acl.
>
> If the second part fails (due to lack of space, for example) and the file
> had no acl attribute to begin with, the system will from now on assume
> that the mask permission bits are actual group permission bits, potentially
> granting access to the wrong users.
>
> Prevent this by only changing the inode mode after the acl has been set.
> Also make reiserfs_xattr_set_handle() return -ENODATA only in case of
> actual error, and return 0 when requested deletion of a nonexistent acl.
> This is more consistent with the behaviour of other *_xattr_set()
> functions and makes the patch simpler.
>
> Signed-off-by: Ernesto A. Fernández <ernesto.mnd.fernandez@...il.com>
Thanks for the patch! I have rebased this onto my tree and merged the
resulting patch (attached).
Honza
> ---
> fs/reiserfs/xattr.c | 4 ++++
> fs/reiserfs/xattr_acl.c | 25 +++++++++----------------
> 2 files changed, 13 insertions(+), 16 deletions(-)
>
> diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c
> index e87aa21..936a57a 100644
> --- a/fs/reiserfs/xattr.c
> +++ b/fs/reiserfs/xattr.c
> @@ -513,6 +513,10 @@ reiserfs_xattr_set_handle(struct reiserfs_transaction_handle *th,
>
> if (!buffer) {
> err = lookup_and_delete_xattr(inode, name);
> + if (flags & XATTR_REPLACE)
> + return err;
> + if (err == -ENODATA)
> + err = 0;
> return err;
> }
>
> diff --git a/fs/reiserfs/xattr_acl.c b/fs/reiserfs/xattr_acl.c
> index 3d2256a..c02e5ac 100644
> --- a/fs/reiserfs/xattr_acl.c
> +++ b/fs/reiserfs/xattr_acl.c
> @@ -237,14 +237,17 @@ __reiserfs_set_acl(struct reiserfs_transaction_handle *th, struct inode *inode,
> void *value = NULL;
> size_t size = 0;
> int error;
> + int update_mode = 0;
> + umode_t mode = inode->i_mode;
>
> switch (type) {
> case ACL_TYPE_ACCESS:
> name = XATTR_NAME_POSIX_ACL_ACCESS;
> if (acl) {
> - error = posix_acl_update_mode(inode, &inode->i_mode, &acl);
> + error = posix_acl_update_mode(inode, &mode, &acl);
> if (error)
> return error;
> + update_mode = 1;
> }
> break;
> case ACL_TYPE_DEFAULT:
> @@ -264,25 +267,15 @@ __reiserfs_set_acl(struct reiserfs_transaction_handle *th, struct inode *inode,
>
> error = reiserfs_xattr_set_handle(th, inode, name, value, size, 0);
>
> - /*
> - * Ensure that the inode gets dirtied if we're only using
> - * the mode bits and an old ACL didn't exist. We don't need
> - * to check if the inode is hashed here since we won't get
> - * called by reiserfs_inherit_default_acl().
> - */
> - if (error == -ENODATA) {
> - error = 0;
> - if (type == ACL_TYPE_ACCESS) {
> + kfree(value);
> + if (!error) {
> + set_cached_acl(inode, type, acl);
> + if (update_mode) {
> + inode->i_mode = mode;
> inode->i_ctime = current_time(inode);
> mark_inode_dirty(inode);
> }
> }
> -
> - kfree(value);
> -
> - if (!error)
> - set_cached_acl(inode, type, acl);
> -
> return error;
> }
>
> --
> 2.1.4
>
>
--
Jan Kara <jack@...e.com>
SUSE Labs, CR
View attachment "0001-reiserfs-preserve-i_mode-if-__reiserfs_set_acl-fails.patch" of type "text/x-patch" (2176 bytes)
Powered by blists - more mailing lists