For any fixed width integer type, the worst case of the dead simple loop (all bits are zero) is a fixed upper bound. So you don't mean "constant bounded" either. You mean something more like "clever C code that usually runs faster than the obvious loop".

See my "if it's not unbearably pedantic" comment earlier ;-) Again, by "primitive" I meant HW-level primitive. I agree that's not wholly clear from what I wrote, but really don't care - it's a trivial point that makes no difference in context. The lack of an integer type in C wide enough to support the division the paper uses is _the_ deal-breaker. That C doesn't define a count-leading-zero function either is just a flea on the tail of that dog.