EqInst related clean up

[ghc-hetmet.git] / compiler / typecheck / TcRnTypes.lhs

diff --git a/compiler/typecheck/TcRnTypes.lhs b/compiler/typecheck/TcRnTypes.lhs
index 6c8d648..ec2f5da 100644 (file)
--- a/compiler/typecheck/TcRnTypes.lhs
+++ b/compiler/typecheck/TcRnTypes.lhs
@@ -3,11 +3,11 @@
 % (c) The GRASP Project, Glasgow University, 1992-2002
 %
 \begin{code}
-{-# OPTIONS_GHC -w #-}
+{-# OPTIONS -w #-}
 -- The above warning supression flag is a temporary kludge.
 -- While working on this module you are encouraged to remove it and fix
 -- any warnings in the module. See
---     http://hackage.haskell.org/trac/ghc/wiki/WorkingConventions#Warnings
+--     http://hackage.haskell.org/trac/ghc/wiki/Commentary/CodingStyle#Warnings
 -- for details
 
 module TcRnTypes(
@@ -635,6 +635,17 @@ In addition to the basic Haskell variants of 'Inst's, they can now also
 represent implication constraints 'forall tvs. (reft, given) => wanted'
 and equality constraints 'co :: ty1 ~ ty2'.
 
+NB: Equalities occur in two flavours:
+
+  (1) Dict {tci_pred = EqPred ty1 ty2}
+  (2) EqInst {tci_left = ty1, tci_right = ty2, tci_co = coe}
+
+The former arises from equalities in contexts, whereas the latter is used
+whenever the type checker introduces an equality (e.g., during deferring
+unification).
+
+I am not convinced that this duplication is necessary or useful! -=chak
+
 \begin{code}
 data Inst
   = Dict {