Repository : ssh://darcs.haskell.org//srv/darcs/ghc

On branch  : master

http://hackage.haskell.org/trac/ghc/changeset/522a1552d0546458bb1c0ff7052175606ac36d99

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

commit 522a1552d0546458bb1c0ff7052175606ac36d99
Author: Simon Peyton Jones <[email protected]>
Date:   Tue Oct 2 08:31:31 2012 +0100

    Comments, and unused import

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

 compiler/typecheck/FamInst.lhs    |    1 -
 compiler/typecheck/TcEvidence.lhs |    9 +++++----
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/compiler/typecheck/FamInst.lhs b/compiler/typecheck/FamInst.lhs
index 2758890..45ef026 100644
--- a/compiler/typecheck/FamInst.lhs
+++ b/compiler/typecheck/FamInst.lhs
@@ -26,7 +26,6 @@ import Name
 import Module
 import Outputable
 import UniqFM
-import VarSet
 import FastString
 import Util
 import Maybes
diff --git a/compiler/typecheck/TcEvidence.lhs 
b/compiler/typecheck/TcEvidence.lhs
index 0c31bc0..b160c32 100644
--- a/compiler/typecheck/TcEvidence.lhs
+++ b/compiler/typecheck/TcEvidence.lhs
@@ -508,16 +508,17 @@ We do quite often need to get a TcCoercion from an 
EvTerm; see
 
 INVARIANT: The evidence for any constraint with type (t1~t2) is 
 a coercion evidence term.  Consider for example
-    [G] g :: F Int a
+    [G] d :: F Int a
 If we have
     ax7 a :: F Int a ~ (a ~ Bool)
 then we do NOT generate the constraint
-    [G} (g |> ax7 a) :: a ~ Bool
-because that does not satisfy the invariant.  Instead we make a binding
+    [G} (d |> ax7 a) :: a ~ Bool
+because that does not satisfy the invariant (d is not a coercion variable).  
+Instead we make a binding
     g1 :: a~Bool = g |> ax7 a
 and the constraint
     [G] g1 :: a~Bool
-See Trac [7238]
+See Trac [7238] and Note [Bind new Givens immediately] in TcSMonad
 
 Note [EvBinds/EvTerm]
 ~~~~~~~~~~~~~~~~~~~~~



_______________________________________________
Cvs-ghc mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/cvs-ghc

Reply via email to