ocfs2/xattr: Fix a bug in xattr allocation estimation
authorTao Ma <tao.ma@oracle.com>
Wed, 19 Nov 2008 08:48:41 +0000 (16:48 +0800)
committerMark Fasheh <mfasheh@suse.com>
Mon, 5 Jan 2009 16:36:55 +0000 (08:36 -0800)
When we extend one xattr's value to a large size, the old value size might
be smaller than the size of a value root. In those cases, we still need to
guess the metadata allocation.

Reported-by: Tiger Yang <tiger.yang@oracle.com>
Signed-off-by: Tao Ma <tao.ma@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
fs/ocfs2/xattr.c

index 8af29b3bd6de21a2358f0050118e2a74ea034fc3..d0b94edb9662684529141afa2bf8455f590a56bd 100644 (file)
@@ -2270,6 +2270,7 @@ static int ocfs2_calc_xattr_set_need(struct inode *inode,
                                                                 value_size);
                        xv = (struct ocfs2_xattr_value_root *)
                             (base + name_offset + name_len);
+                       value_size = OCFS2_XATTR_ROOT_SIZE;
                } else
                        xv = &def_xv.xv;
 
@@ -2283,7 +2284,8 @@ static int ocfs2_calc_xattr_set_need(struct inode *inode,
                                                             &xv->xr_list,
                                                             new_clusters -
                                                             old_clusters);
-                       goto out;
+                       if (value_size >= OCFS2_XATTR_ROOT_SIZE)
+                               goto out;
                }
        } else {
                /*