RE: Yet another External Core bug

| The attached file (generated by running ghc -fext-core on the Fibheaps
| benchmark from the nofib suite) fails to typecheck:
| 
| $ ghc -dcore-lint Fibheaps.hcr
| 
| Couldn't match `#' against `*'
|     Expected kind: #
|     Inferred kind: *
| When checking kinds in `GHC.Prim.(->) (GHC.Prim.State# s)'

I've found what's going on here.  Here is the comment from the
definition of the type constructor for (->):

funTyCon = mkFunTyCon funTyConName 
                (mkArrowKinds [liftedTypeKind, liftedTypeKind]
liftedTypeKind)
        -- You might think that (->) should have type (? -> ? -> *), and
you'd be right
        -- But if we do that we get kind errors when saying
        --      instance Control.Arrow (->)
        -- becuase the expected kind is (*->*->*).  The trouble is that
the
        -- expected/actual stuff in the unifier does not go
contra-variant, whereas
        -- the kind sub-typing does.  Sigh.  It really only matters if
you use (->) in
        -- a prefix way, thus:  (->) Int# Int#.  And this is unusual.


In short, it's really a bug, but not a particularly easy one to fix.
Rumination required.  But it probably only happens on a few programs,
right?

S