new SSA IL

[Matthew Fluet]

> If  goto's  pass  arguments,  is  their semantics assignment to the argument?
> I.e., I am in a `function', x is in scope (it is the piece of the list  I  am
> working  on),  I  call  myself  with  the tail of x as an arguemnt.  Thus you
> really do have mutation of variables, right?

Sort of.  Maybe it's better to think of it as mutation of a portion of the
environment.  The semantics of the SSA IL (minus the exceptions) would, I
think, be pretty much identical to the semantics of the CPS IL.  If you
look at the definition of the semantics in the contification paper, you'll
notice that we never care about the lexical scoping.

> It is amusing that the whole thing some what reminds me of the  ICFP  contest
> and the realization that block structuring is bad from the point of view of a
> low-level implementation.

;)  This thought struck me as well.