[project @ 2001-08-23 08:43:30 by simonpj]

-----------------------------------
	Correct a horrible error in repType
	-----------------------------------

repType is meant to give the underlying representation of a type.
But it wasn't taking account of the fact that *recursive* newtypes are
still represented by a TyConApp.  (Non-recursive ones behave much more
like type synonyms now.)

As a result, if we have

	newtype F = F (F->F)

then Bad Things happen if we try to seq x::F.  We decide whether to
push an ordinary return address or a SEQ frame based on the type,
and repType didn't expose the fact that F is represented by a function type.

Aargh. codeGen/should_run/cg050 now tests for this.

diff --git a/ghc/compiler/types/Type.lhs b/ghc/compiler/types/Type.lhs
index be39f10..33cd4b1 100644 (file)
--- a/ghc/compiler/types/Type.lhs
+++ b/ghc/compiler/types/Type.lhs
@@ -398,15 +398,22 @@ repType looks through
        (b) synonyms
        (c) predicates
        (d) usage annotations
+       (e) [recursive] newtypes
 It's useful in the back end.
 
+Remember, non-recursive newtypes get expanded as part of the SourceTy case,
+but recursive ones are represented by TyConApps and have to be expanded
+by steam.
+
 \begin{code}
 repType :: Type -> Type
-repType (ForAllTy _ ty) = repType ty
-repType (NoteTy   _ ty) = repType ty
-repType (SourceTy  p)   = repType (sourceTypeRep p)
-repType (UsageTy  _ ty) = repType ty
-repType ty             = ty
+repType (ForAllTy _ ty)   = repType ty
+repType (NoteTy   _ ty)   = repType ty
+repType (SourceTy  p)     = repType (sourceTypeRep p)
+repType (UsageTy  _ ty)   = repType ty
+repType (TyConApp tc tys) | isNewTyCon tc && length tys == tyConArity tc
+                         = repType (newTypeRep tc tys)
+repType ty               = ty
 
 splitRepFunTys :: Type -> ([Type], Type)
 -- Like splitFunTys, but looks through newtypes and for-alls