diff options
| author | Brian Wignall <brianwignall@gmail.com> | 2021-02-05 09:03:21 -0500 |
|---|---|---|
| committer | Marge Bot <ben+marge-bot@smart-cactus.org> | 2021-02-06 09:30:34 -0500 |
| commit | 3da472f0e78fe5f1068be0cc2b1c0762532da9f9 (patch) | |
| tree | d88a147c49aa25fff13866de9b87c712213fbd3f /compiler/GHC/Tc/Gen/App.hs | |
| parent | 9b7dcd80f16efda57e4e51e39d61b55a11c72014 (diff) | |
| download | haskell-3da472f0e78fe5f1068be0cc2b1c0762532da9f9.tar.gz | |
Fix typos
Diffstat (limited to 'compiler/GHC/Tc/Gen/App.hs')
| -rw-r--r-- | compiler/GHC/Tc/Gen/App.hs | 6 |
1 files changed, 3 insertions, 3 deletions
diff --git a/compiler/GHC/Tc/Gen/App.hs b/compiler/GHC/Tc/Gen/App.hs index 83f8cc567d..5b34952d65 100644 --- a/compiler/GHC/Tc/Gen/App.hs +++ b/compiler/GHC/Tc/Gen/App.hs @@ -291,7 +291,7 @@ tcApp rn_expr exp_res_ty -- Zonk the result type, to ensure that we substitute out -- any filled-in instantiation variable before calling tcWrapResultMono - -- In the Check case, this isn't really necessary, becuase tcWrapResultMono + -- In the Check case, this isn't really necessary, because tcWrapResultMono -- just drops to tcUnify; but in the Infer case a filled-in instantiation -- variable might perhaps escape into the constraint generator. The safe -- thing to do is to any instantaition variables away. @@ -318,7 +318,7 @@ wantQuickLook _ = xoptM LangExt.Impredi zonkQuickLook :: Bool -> TcType -> TcM TcType -- After all Quick Look unifications are done, zonk to ensure that all --- instantation variables are substituted away +-- instantiation variables are substituted away -- -- So far as the paper is concerned, this step applies -- the poly-substitution Theta, learned by QL, so that we @@ -677,7 +677,7 @@ over any visible type arguments. We've already inferred the type of the function (in tcInferAppHead), so we'll /already/ have emitted a Hole constraint; failing preserves that constraint. -We do /not/ want to fail altogether in this case (via failM) becuase +We do /not/ want to fail altogether in this case (via failM) because that may abandon an entire instance decl, which (in the presence of -fdefer-type-errors) leads to leading to #17792. |