Mark Fasheh wrote: > On Fri, Jun 13, 2008 at 09:48:24AM +0800, Tao Ma wrote: >>> Hmm, at this point, wouldn't it make sense to have a couple high-level >>> "ocfs2_foo_insert_extent" functions whcih build up anm ocfs2_extent_tree >>> and >>> then pass it down to the common ocfs2_insert_extent? >> Why do we need that? Just to reduce the parameter to one >> "ocfs2_extent_tree *"? ;) Any other benefit? > > No, that's mostly what I was concerned with. Well, that and type-checking - > we don't have to have a void * in ocfs2_insert_extent any more. Then how about add 3 functions: ocfs2_inode_insert_extent, ocfs2_xattr_value_insert_extent and ocfs2_xattr_tree_insert_extent(for xattr index btree, and it will show up in the last patch). All the old callers in file.c etc will call ocfs2_dinode_insert_extent, while the other two handle the xattr issue. And the init of extent tree are handled by these functions?
Regards, Tao _______________________________________________ Ocfs2-devel mailing list [email protected] http://oss.oracle.com/mailman/listinfo/ocfs2-devel
