Jaakko Heinonen <jh at FreeBSD.org> writes: > This assumes that sizeof(*i) is always sizeof(long) or > sizeof(uintmax_t). I prefer the fix bde suggested in another mail if > it's possible (I don't know if *i is always type of register_t). It is. I didn't realize there was a u_register_t. DES -- Dag-Erling Smørgrav - des at des.no