[Git][ghc/ghc][wip/T25647] enhance trace logging in tcConArg and cloneAnonMetaTyVar for better debugging

[Patrick (@soulomoon)]

Patrick pushed to branch wip/T25647 at Glasgow Haskell Compiler / GHC


Commits:
d8258877 by Patrick at 2025-03-05T06:19:27+08:00
enhance trace logging in tcConArg and cloneAnonMetaTyVar for better debugging

- - - - -


3 changed files:

- compiler/GHC/Tc/TyCl.hs
- compiler/GHC/Tc/TyCl/Instance.hs
- compiler/GHC/Tc/Utils/TcMType.hs


Changes:

=====================================
compiler/GHC/Tc/TyCl.hs
=====================================
@@ -3409,9 +3409,10 @@ without treating the explicitly-quantified ones specially. Wrinkles:
     In step 1 we do /not/ want to get
        newtype instance forall r .  Fix2 (f :: TYPE r -> TYPE r) :: TYPE r where
     If we do, we'll get that same "newtype must not be GADT" error as for N above.
-    Rather, we want to default the RuntimeRep variable r := LiftedRep. The key thing
-    is that we must make the /same/ choice here as we do in kind-checking the data
-    constructor's type.
+    Rather, we want to default the RuntimeRep variable r := LiftedRep. See the call
+    to `quantifyTyVars` in `tcDataFamInstHeader`. The key thing is that we must make
+    the /same/ choice here as we do in kind-checking the data constructor's type
+    in `kindGeneralizeAll` in `tcConDecl`.
 
 See also Note [Re-quantify type variables in rules] in GHC.Tc.Gen.Rule, which
 explains a /very/ similar design when generalising over the type of a rewrite
@@ -4034,10 +4035,10 @@ tcConArg :: ConArgKind   -- expected kind for args; always OpenKind for datatype
                          -- but might be an unlifted type with UnliftedNewtypes
          -> HsScaled GhcRn (LHsType GhcRn) -> TcM (Scaled TcType, HsSrcBang)
 tcConArg exp_kind (HsScaled w bty)
-  = do  { traceTc "tcConArg 1" (ppr bty)
+  = do  { traceTc "tcConArg 1: " (ppr bty <+> ppr exp_kind)
         ; arg_ty <- tcCheckLHsTypeInContext (getBangType bty) exp_kind
         ; w' <- tcDataConMult w
-        ; traceTc "tcConArg 2" (ppr bty)
+        ; traceTc "tcConArg 2: " (ppr bty <+> ppr arg_ty)
         ; return (Scaled w' arg_ty, getBangStrictness bty) }
 
 tcRecConDeclFields :: ConArgKind


=====================================
compiler/GHC/Tc/TyCl/Instance.hs
=====================================
@@ -977,6 +977,8 @@ tcDataFamInstHeader mb_clsinfo skol_info fam_tc hs_outer_bndrs fixity
        -- See GHC.Tc.TyCl Note [Generalising in tcTyFamInstEqnGuts]
        ; dvs  <- candidateQTyVarsWithBinders outer_tvs lhs_ty
        ; qtvs <- quantifyTyVars skol_info dvs
+                 -- Have to make a same defaulting choice for reuslt kind here
+                 -- and the `kindGeneralizeAll` in `tcConDecl`.
                  -- see (GT4) in
                  -- GHC.Tc.TyCl Note [Generalising in tcTyFamInstEqnGuts]
 


=====================================
compiler/GHC/Tc/Utils/TcMType.hs
=====================================
@@ -827,7 +827,7 @@ cloneAnonMetaTyVar info tv kind
   = do  { details <- newMetaDetails info
         ; name    <- cloneMetaTyVarName (tyVarName tv)
         ; let tyvar = mkTcTyVar name kind details
-        ; traceTc "cloneAnonMetaTyVar" (ppr tyvar <+> dcolon <+> ppr (tyVarKind tyvar))
+        ; traceTc "cloneAnonMetaTyVar" (ppr tyvar <+> dcolon <+> ppr (tyVarKind tyvar) <+> text "from" <+> ppr tv)
         ; return tyvar }
 
 -- Make a new CycleBreakerTv. See Note [Type equality cycles]



View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/d82588777cc57efd47036da3568197616cf07908

-- 
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/d82588777cc57efd47036da3568197616cf07908
You're receiving this email because of your account on gitlab.haskell.org.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.haskell.org/pipermail/ghc-commits/attachments/20250304/ac8ee0d0/attachment-0001.html>