[commit: ghc] master: Comments only (2140e8a)

[git at git.haskell.org]

Repository : ssh://git@git.haskell.org/ghc

On branch  : master
Link       : http://ghc.haskell.org/trac/ghc/changeset/2140e8a5ce58bc619cdbeea0fb93b07779964b4f/ghc

>---------------------------------------------------------------

commit 2140e8a5ce58bc619cdbeea0fb93b07779964b4f
Author: Simon Peyton Jones <simonpj at microsoft.com>
Date:   Thu Oct 10 10:53:41 2013 +0100

    Comments only


>---------------------------------------------------------------

2140e8a5ce58bc619cdbeea0fb93b07779964b4f
 compiler/typecheck/TcInstDcls.lhs |    3 +++
 1 file changed, 3 insertions(+)

diff --git a/compiler/typecheck/TcInstDcls.lhs b/compiler/typecheck/TcInstDcls.lhs
index 381c082..6d4e620 100644
--- a/compiler/typecheck/TcInstDcls.lhs
+++ b/compiler/typecheck/TcInstDcls.lhs
@@ -1358,6 +1358,9 @@ tcInstanceMethods dfun_id clas tyvars dfun_ev_vars inst_tys
 -- Notice that the dictionary bindings "..d1..d2.." must be generated
 -- by the constraint solver, since the <context> may be
 -- user-specified.
+--
+-- See also Note [Newtype deriving superclasses] in TcDeriv
+-- for why we don't just coerce the superclass
 
   = do { rep_d_stuff <- checkConstraints InstSkol tyvars dfun_ev_vars $
                         emitWanted ScOrigin rep_pred