[commit: ghc] master: Comments (2e4f364)
git at git.haskell.org
git at git.haskell.org
Fri Sep 26 13:42:53 UTC 2014
Repository : ssh://git@git.haskell.org/ghc
On branch : master
Link : http://ghc.haskell.org/trac/ghc/changeset/2e4f36425320f21ebc6039790034df20f2d470f9/ghc
>---------------------------------------------------------------
commit 2e4f36425320f21ebc6039790034df20f2d470f9
Author: Simon Peyton Jones <simonpj at microsoft.com>
Date: Fri Sep 26 14:42:40 2014 +0100
Comments
>---------------------------------------------------------------
2e4f36425320f21ebc6039790034df20f2d470f9
compiler/typecheck/TcSimplify.lhs | 1 +
compiler/typecheck/TcUnify.lhs | 4 ++--
2 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/compiler/typecheck/TcSimplify.lhs b/compiler/typecheck/TcSimplify.lhs
index ceb517d..d834763 100644
--- a/compiler/typecheck/TcSimplify.lhs
+++ b/compiler/typecheck/TcSimplify.lhs
@@ -364,6 +364,7 @@ simplifyInfer _top_lvl apply_mr name_taus wanteds
else do
{ -- Step 7) Emit an implication
+ -- See Trac #9633 for an instructive example
let minimal_flat_preds = mkMinimalBySCs bound
-- See Note [Minimize by Superclasses]
skol_info = InferSkol [ (name, mkSigmaTy [] minimal_flat_preds ty)
diff --git a/compiler/typecheck/TcUnify.lhs b/compiler/typecheck/TcUnify.lhs
index b66f06b..389c4a3 100644
--- a/compiler/typecheck/TcUnify.lhs
+++ b/compiler/typecheck/TcUnify.lhs
@@ -1071,9 +1071,9 @@ We must use the careful function lookupTcTyVar to see if a kind
variable is filled or unifiable. It checks for touchablity, and kind
variables can certainly be untouchable --- for example the variable
might be bound outside an enclosing existental pattern match that
-binds an inner kind variable, which we don't want ot escape outside.
+binds an inner kind variable, which we don't want to escape outside.
-This, or something closely related, was teh cause of Trac #8985.
+This, or something closely related, was the cause of Trac #8985.
Note [Unifying kind variables]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
More information about the ghc-commits
mailing list