GFS2: Fix very unlikley memory leak in ACL xattr code

This was spotted by automated code analysis. In case reading
an ACL xattr failed (only likely to happen if there is an I/O
error for example, and even then only with unstuffed xattrs,
so pretty difficult to trigger) a small amount of memory could
potentially be leaked.

This patch adds a kfree to the error path, and also removes a
test which is no longer required (gfs2_ea_get_copy always
returns either a negative error, or a length)

Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>

diff --git a/fs/gfs2/xattr.c b/fs/gfs2/xattr.c
index 71d7bf830c09a44ed22884c75c5ae3f77ec1824d..a201a1d57a6d3d30c39310230fc92a9266f50c4f 100644 (file)
--- a/fs/gfs2/xattr.c
+++ b/fs/gfs2/xattr.c
@@ -549,9 +549,10 @@ int gfs2_xattr_acl_get(struct gfs2_inode *ip, const char *name, char **ppdata)
                goto out;
 
        error = gfs2_ea_get_copy(ip, &el, data, len);
-       if (error == 0)
-               error = len;
-       *ppdata = data;
+       if (error < 0)
+               kfree(data);
+       else
+               *ppdata = data;
 out:
        brelse(el.el_bh);
        return error;