[racket] A primitive more fundamental than a continuation?

Galler schrieb:
> 'Racket could implement mutable continuations, but has not'

I was just thinking whether mutable continuations can
be used for some JIT-ing. Had already a question about
call-site code replacement here. And there some response
was to use set! on a symbol. But using set-car! looks
closer to what I would need.

But it would be more a look-ahead JIT-ing. Since when
inside the continuation, it would be already too late
I guess. The continuation would be not that aware of the
call-site. Or maybe via some inspection also some handles
can be be acquired.

Hm...