panic! (the `impossible' happened)

[Simon Peyton-Jones]

|   $ ghc -c Formula.hs -fallow-overlapping-instances \
|                       -fallow-undecidable-instances -fglasgow-exts
| 
|   panic! (the `impossible' happened):
|           applyTy

Excellent bug!  Thank you.  

Turns out that it happens when there are join points combined
with existential constructors.  The latter came along after the former
were implemented.

Solution: replace the tail end of Simplify.mkDupableAlt with the code below.

Simon M is going to commit this to the ghc-407-banch of the repository,
but I don't know if he's done so yet.

It'll be fixed in 4.08.2, which we'll put out soon.

QUESTION FOR EVERYONE

	Is anyone stuck becuase of a bug in 4.08.1 that they
	can't work around by changing the source or slurping the
	draft 4.08.2 from teh repository?

Simon

================ STUFF TO PUT AT END OF Simplify.lhs

	-- See comment about "$j" name above
    newId SLIT("$j") (foldr mkPiType rhs_ty' final_bndrs')	$ \
join_bndr ->
	-- Notice the funky mkPiType.  If the contructor has existentials
	-- it's possible that the join point will be abstracted over
	-- type varaibles as well as term variables.
	--  Example:  Suppose we have
	--	data T = forall t.  C [t]
	--  Then faced with
	--	case (case e of ...) of
	--	    C t xs::[t] -> rhs
	--  We get the join point
	--	let j :: forall t. [t] -> ...
	--	    j = /\t \xs::[t] -> rhs
	--	in
	--	case (case e of ...) of
	--	    C t xs::[t] -> j t xs

    let 
	-- We make the lambdas into one-shot-lambdas.  The
	-- join point is sure to be applied at most once, and doing so
	-- prevents the body of the join point being floated out by
	-- the full laziness pass
	really_final_bndrs = map one_shot final_bndrs'
	one_shot v | isId v    = setOneShotLambda v
		   | otherwise = v
    in
    returnSmpl ([NonRec join_bndr (mkLams really_final_bndrs rhs')],
		(con, bndrs, mkApps (Var join_bndr) final_args))