- ghc-commits - Haskell.org

[Git][ghc/ghc][wip/spj-apporv-Oct24] hopefully fixes all the error messages
by Apoorv Ingle (＠ani) 25 Nov '25

25 Nov '25

Apoorv Ingle pushed to branch wip/spj-apporv-Oct24 at Glasgow Haskell Compiler / GHC Commits: f42facf3 by Apoorv Ingle at 2025-11-25T17:36:39-06:00 hopefully fixes all the error messages - - - - - 2 changed files: - compiler/GHC/Tc/Gen/Expr.hs - compiler/GHC/Tc/Gen/Match.hs Changes: ===================================== compiler/GHC/Tc/Gen/Expr.hs ===================================== @@ -335,7 +335,7 @@ tcExpr e@(HsLit x lit) res_ty ; tcWrapResult e (HsLit x (convertLit lit)) lit_ty res_ty } tcExpr (HsPar x expr) res_ty - = do { expr' <- tcMonoLExpr expr res_ty + = do { expr' <- tcMonoLExprNC expr res_ty ; return (HsPar x expr') } tcExpr (HsPragE x prag expr) res_ty ===================================== compiler/GHC/Tc/Gen/Match.hs ===================================== @@ -42,7 +42,7 @@ import GHC.Prelude import {-# SOURCE #-} GHC.Tc.Gen.Expr( tcSyntaxOp, tcInferRho, tcInferRhoFRRNC , tcMonoLExprNC, tcMonoLExpr, tcExpr , tcCheckMonoExpr, tcCheckMonoExprNC - , tcCheckPolyExpr, tcPolyExpr ) + , tcCheckPolyExpr, tcPolyLExpr ) import GHC.Rename.Utils ( bindLocalNames ) import GHC.Tc.Errors.Types @@ -420,15 +420,10 @@ tcDoStmts mDoExpr ss@(L _ stmts) res_ty } tcBody :: LHsExpr GhcRn -> ExpRhoType -> TcM (LHsExpr GhcTc) -tcBody (L lspan e) res_ty - = L lspan <$> - do { traceTc "tcBody" (ppr res_ty) - ; setSrcSpanA lspan $ - addErrCtxt (ExprCtxt e) $ - -- We want the right hand side of a match or an equation - -- to always get printed in the error context - tcPolyExpr e res_ty - } +tcBody body res_ty + = do { traceTc "tcBody" (ppr res_ty) + ; tcPolyLExpr body res_ty + } {- ************************************************************************ View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/f42facf387a0ee107782baf225a9b6b… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/f42facf387a0ee107782baf225a9b6b… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/spj-apporv-Oct24] accept a test case. Remove HsPar as a special case from add_expr_ctxt
by Apoorv Ingle (＠ani) 25 Nov '25

25 Nov '25

Apoorv Ingle pushed to branch wip/spj-apporv-Oct24 at Glasgow Haskell Compiler / GHC Commits: ac1c8658 by Apoorv Ingle at 2025-11-25T16:57:11-06:00 accept a test case. Remove HsPar as a special case from add_expr_ctxt - - - - - 3 changed files: - compiler/GHC/Tc/Gen/Head.hs - compiler/GHC/Tc/Gen/Match.hs - testsuite/tests/indexed-types/should_fail/T1897b.stderr Changes: ===================================== compiler/GHC/Tc/Gen/Head.hs ===================================== @@ -1091,14 +1091,8 @@ add_expr_ctxt e thing_inside -- when we don't want to say "In the expression: _", -- because it is mentioned in the error message itself - HsPar{} -> thing_inside - -- We don't want to say 'In the expression (e)', - -- we just want to say 'In the expression, 'e' - -- which will be handeled by the recursive call in thing_inside - ExprWithTySig _ (L _ e') _ - | XExpr (ExpandedThingRn o _) <- e' -> addExpansionErrCtxt o (ExprCtxt e) - thing_inside + | XExpr (ExpandedThingRn o _) <- e' -> addExpansionErrCtxt o (ExprCtxt e) thing_inside -- There is a special case for expressions with signatures to avoid having too verbose -- error context. So here we flip the ErrCtxt state to expanded if the expression is expanded. -- c.f. RecordDotSyntaxFail9 ===================================== compiler/GHC/Tc/Gen/Match.hs ===================================== @@ -426,7 +426,7 @@ tcBody (L lspan e) res_ty ; setSrcSpanA lspan $ addErrCtxt (ExprCtxt e) $ -- We want the right hand side of a match or an equation - -- to always get printed + -- to always get printed in the error context tcPolyExpr e res_ty } ===================================== testsuite/tests/indexed-types/should_fail/T1897b.stderr ===================================== @@ -1,14 +1,14 @@ T1897b.hs:16:1: error: [GHC-83865] - • Couldn't match type: Depend b0 - with: Depend b - Expected: t (Depend b) -> Bool - Actual: t (Depend b0) -> Bool + • Couldn't match type: Depend a0 + with: Depend a + Expected: t (Depend a) -> Bool + Actual: t (Depend a0) -> Bool Note: ‘Depend’ is a non-injective type family. - The type variable ‘b0’ is ambiguous + The type variable ‘a0’ is ambiguous • In the ambiguity check for the inferred type for ‘isValid’ To defer the ambiguity check to use sites, enable AllowAmbiguousTypes When checking the inferred type - isValid :: forall {t :: * -> *} {b}. - (Foldable t, Bug b) => - t (Depend b) -> Bool + isValid :: forall {t :: * -> *} {a}. + (Foldable t, Bug a) => + t (Depend a) -> Bool View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/ac1c86588be8d88f4fe08c520960555… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/ac1c86588be8d88f4fe08c520960555… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/amg/castz] 4 commits: Accept output change for prog-mhu002c
by Adam Gundry (＠adamgundry) 25 Nov '25

25 Nov '25

Adam Gundry pushed to branch wip/amg/castz at Glasgow Haskell Compiler / GHC Commits: f9292660 by Adam Gundry at 2025-11-25T14:43:11+00:00 Accept output change for prog-mhu002c - - - - - b5ebc105 by Adam Gundry at 2025-11-25T19:15:27+00:00 Fix forall visibility bugs - - - - - 6aa2899f by Adam Gundry at 2025-11-25T19:32:01+00:00 Optimise ZCoercion in simplifier - - - - - 5afbb4f8 by Adam Gundry at 2025-11-25T20:48:39+00:00 Experiment with heuristic so zonker zaps only large casts - - - - - 5 changed files: - compiler/GHC/Core/Coercion.hs - compiler/GHC/Core/Opt/Arity.hs - compiler/GHC/Core/Opt/Simplify/Iteration.hs - compiler/GHC/Tc/Zonk/Type.hs - testsuite/tests/ghci/prog-mhu002/prog-mhu002c.stdout Changes: ===================================== compiler/GHC/Core/Coercion.hs ===================================== @@ -984,12 +984,13 @@ mkForAllCo v visL visR kind_co co | otherwise = mk_forall_co v visL visR kind_co co -mkForAllCastCo :: HasDebugCallStack => TyCoVar -> ForAllTyFlag -> ForAllTyFlag - -> CastCoercion -> CastCoercion -mkForAllCastCo v visL visR cco = case cco of +mkForAllCastCo :: HasDebugCallStack => Role -> TyCoVar -> ForAllTyFlag -> ForAllTyFlag + -> Type -> CastCoercion -> CastCoercion +mkForAllCastCo r v visL visR ty cco = case cco of CCoercion co -> CCoercion (mkForAllCo v visL visR MRefl co) - ZCoercion ty cos -> ZCoercion (mkTyCoForAllTy v visL ty) cos - ReflCastCo -> ReflCastCo + ZCoercion ty cos -> ZCoercion (mkTyCoForAllTy v visR ty) cos + ReflCastCo | visL `eqForAllVis` visR -> ReflCastCo + | otherwise -> CCoercion (mk_forall_co v visL visR MRefl (mkReflCo r ty)) -- mkForAllVisCos [tv{vis}] constructs a cast -- forall tv. res ~R# forall tv{vis} res`. @@ -1816,7 +1817,7 @@ mkPiCastCo :: Role -> Var -> CastCoercion -> CastCoercion mkPiCastCo _ _ ReflCastCo = ReflCastCo mkPiCastCo r v (CCoercion co) = CCoercion (mkPiCo r v co) mkPiCastCo _ v (ZCoercion ty cos) - | isTyVar v = ZCoercion (mkForAllTy (Bndr v vis) ty) cos + | isTyVar v = ZCoercion (mkTyCoForAllTy v vis ty) cos | otherwise = ZCoercion (mkFunctionType (idMult v) (varType v) ty) cos where vis = coreTyLamForAllTyFlag ===================================== compiler/GHC/Core/Opt/Arity.hs ===================================== @@ -2359,12 +2359,12 @@ mkEtaWW orig_oss ppr_orig_expr in_scope orig_ty -- with an explicit lambda having a non-function type mkEtaForAllMCo :: ForAllTyBinder -> Type -> CastCoercion -> CastCoercion -mkEtaForAllMCo bdnr@(Bndr tcv vis) ty mco +mkEtaForAllMCo (Bndr tcv vis) ty mco = case mco of ReflCastCo | vis == coreTyLamForAllTyFlag -> ReflCastCo | otherwise -> mk_fco (mkRepReflCo ty) CCoercion co -> mk_fco co - ZCoercion tyR cos -> ZCoercion (mkForAllTy bdnr tyR) cos + ZCoercion tyR cos -> ZCoercion (mkTyCoForAllTy tcv coreTyLamForAllTyFlag tyR) cos where mk_fco co = CCoercion (mkForAllCo tcv vis coreTyLamForAllTyFlag MRefl co) -- coreTyLamForAllTyFlag: See Note [The EtaInfo mechanism], particularly @@ -2723,7 +2723,7 @@ tryEtaReduce rec_ids bndrs body eval_sd -- Float app ticks: \x -> Tick t (e x) ==> Tick t e go (b : bs) (App fun arg) co - | Just (co', ticks) <- ok_arg b arg co (exprType fun) + | Just (co', ticks) <- ok_arg b arg co (exprType fun) (exprType (App fun arg)) = fmap (flip (foldr mkTick) ticks) $ go bs fun co' -- Float arg ticks: \x -> e (Tick t x) ==> Tick t e @@ -2798,15 +2798,16 @@ tryEtaReduce rec_ids bndrs body eval_sd -> CastCoercion -- Of kind (t1~t2) -> Type -- Type (arg_t -> t1) of the function -- to which the argument is supplied + -> Type -- Type t1 of the result (AMG TODO: avoid needing to pass this?) -> Maybe (CastCoercion -- Of type (arg_t -> t1 ~ bndr_t -> t2) -- (and similarly for tyvars, coercion args) , [CoreTickish]) -- See Note [Eta reduction with casted arguments] - ok_arg bndr (Type arg_ty) co fun_ty + ok_arg bndr (Type arg_ty) co fun_ty res_ty | Just tv <- getTyVar_maybe arg_ty , bndr == tv = case splitForAllForAllTyBinder_maybe fun_ty of Just (Bndr _ vis, _) -> Just (fco, []) - where !fco = mkForAllCastCo tv vis coreTyLamForAllTyFlag co + where !fco = mkForAllCastCo Representational tv vis coreTyLamForAllTyFlag res_ty co -- The lambda we are eta-reducing always has visibility -- 'coreTyLamForAllTyFlag' which may or may not match -- the visibility on the inner function (#24014) @@ -2814,24 +2815,24 @@ tryEtaReduce rec_ids bndrs body eval_sd (text "fun:" <+> ppr bndr $$ text "arg:" <+> ppr arg_ty $$ text "fun_ty:" <+> ppr fun_ty) - ok_arg bndr (Var v) co fun_ty + ok_arg bndr (Var v) co fun_ty _ | bndr == v , let mult = idMult bndr , Just (_af, fun_mult, _, _) <- splitFunTy_maybe fun_ty , mult `eqType` fun_mult -- There is no change in multiplicity, otherwise we must abort = Just (mkFunResCastCo Representational bndr co, []) - ok_arg bndr (Cast e co_arg) co fun_ty + ok_arg bndr (Cast e co_arg) co fun_ty _ | (ticks, Var v) <- stripTicksTop tickishFloatable e , Just (_, fun_mult, _, res_ty) <- splitFunTy_maybe fun_ty , bndr == v , fun_mult `eqType` idMult bndr - = Just (mkFunCastCoNoFTF Representational fun_mult (castCoercionRKind (exprType e) co_arg) (mkSymCastCo (exprType e) co_arg) res_ty co, ticks) -- TODO check types + = Just (mkFunCastCoNoFTF Representational fun_mult (castCoercionRKind (exprType e) co_arg) (mkSymCastCo (exprType e) co_arg) res_ty co, ticks) -- The simplifier combines multiple casts into one, -- so we can have a simple-minded pattern match here - ok_arg bndr (Tick t arg) co fun_ty - | tickishFloatable t, Just (co', ticks) <- ok_arg bndr arg co fun_ty + ok_arg bndr (Tick t arg) co fun_ty res_ty + | tickishFloatable t, Just (co', ticks) <- ok_arg bndr arg co fun_ty res_ty = Just (co', t:ticks) - ok_arg _ _ _ _ = Nothing + ok_arg _ _ _ _ _ = Nothing -- | Can we eta-reduce the given function -- See Note [Eta reduction soundness], criteria (B), (J), and (W). ===================================== compiler/GHC/Core/Opt/Simplify/Iteration.hs ===================================== @@ -28,7 +28,7 @@ import GHC.Core.Make ( FloatBind, mkImpossibleExpr, castBottomExpr ) import qualified GHC.Core.Make import GHC.Core.Coercion hiding ( substCo, substCoVar ) import GHC.Core.Reduction -import GHC.Core.Coercion.Opt ( optCoercion ) +import GHC.Core.Coercion.Opt ( optCoercion, optCastCoercion ) import GHC.Core.FamInstEnv ( FamInstEnv, topNormaliseType_maybe ) import GHC.Core.DataCon import GHC.Core.Opt.Stats ( Tick(..) ) @@ -1545,7 +1545,7 @@ rebuild_go env expr cont -> rebuild_go env (mkCastCo expr co') cont -- NB: mkCast implements the (Coercion co |> g) optimisation where - co' = optOutCastCoercion env co opt + co' = optOutCoercion env (exprType expr) co opt Select { sc_bndr = bndr, sc_alts = alts, sc_env = se, sc_cont = cont } -> rebuildCase (se `setInScopeFromE` env) expr bndr alts cont @@ -1674,17 +1674,11 @@ on each successive composition -- that's at least quadratic. So: -} -optOutCastCoercion :: SimplEnvIS -> OutCastCoercion -> Bool -> OutCastCoercion -optOutCastCoercion env cco already_optimised = case cco of - ReflCastCo -> ReflCastCo - CCoercion co -> CCoercion (optOutCoercion env co already_optimised) - ZCoercion{} -> cco - -optOutCoercion :: SimplEnvIS -> OutCoercion -> Bool -> OutCoercion +optOutCoercion :: SimplEnvIS -> Type -> OutCastCoercion -> Bool -> OutCastCoercion -- See Note [Avoid re-simplifying coercions] -optOutCoercion env co already_optimised +optOutCoercion env ty co already_optimised | already_optimised = co -- See Note [Avoid re-simplifying coercions] - | otherwise = optCoercion opts empty_subst co + | otherwise = optCastCoercion opts empty_subst ty co where empty_subst = mkEmptySubst (seInScope env) opts = seOptCoercionOpts env @@ -1732,7 +1726,7 @@ simplCast env body co0 cont0 , sc_dup = dup, sc_cont = tail , sc_hole_ty = fun_ty }) | not co_is_opt -- pushCoValArg duplicates the coercion, so optimise first - = addCoerce tyL (optOutCastCoercion (zapSubstEnv env) co co_is_opt) True cont + = addCoerce tyL (optOutCoercion (zapSubstEnv env) tyL co co_is_opt) True cont | Just (_, m_co1, res_ty, m_co2) <- pushCastCoValArg tyL co = {-#SCC "addCoerce-pushCoValArg" #-} @@ -3886,7 +3880,7 @@ mkDupableContWithDmds _ _ (Stop {}) = panic "mkDupableCont" -- Handled by pr mkDupableContWithDmds env dmds (CastIt { sc_co = co, sc_hole_ty = ty, sc_opt = opt, sc_cont = cont }) = do { (floats, cont') <- mkDupableContWithDmds env dmds cont - ; return (floats, CastIt { sc_co = optOutCastCoercion env co opt + ; return (floats, CastIt { sc_co = optOutCoercion env ty co opt , sc_hole_ty = ty , sc_opt = True, sc_cont = cont' }) } -- optOutCoercion: see Note [Avoid re-simplifying coercions] ===================================== compiler/GHC/Tc/Zonk/Type.hs ===================================== @@ -1862,9 +1862,10 @@ zonkEvTerm (EvExpr e) = EvExpr <$> zonkCoreExpr e zonkEvTerm (EvCastExpr e (CCoercion co) co_res_ty) = do { zap_casts <- hasZapCasts <$> lift getDynFlags - ; co_res_ty' <- zonkTcTypeToTypeX co_res_ty - ; if zap_casts - then EvCastExpr <$> zonkCoreExpr e <*> (ZCoercion co_res_ty' <$> zonkShallowCoVarsOfCo co) <*> pure co_res_ty' + ; if zap_casts && coercionSize co > typeSize co_res_ty -- AMG TODO: experimental heuristic + then do { co_res_ty' <- zonkTcTypeToTypeX co_res_ty + ; EvCastExpr <$> zonkCoreExpr e <*> (ZCoercion co_res_ty' <$> zonkShallowCoVarsOfCo co) <*> pure co_res_ty' + } else EvExpr <$> zonkCoreExpr (Cast e (CCoercion co)) } zonkEvTerm ev@(EvCastExpr _ (ZCoercion{}) _) ===================================== testsuite/tests/ghci/prog-mhu002/prog-mhu002c.stdout ===================================== @@ -15,6 +15,7 @@ other dynamic, non-language, flag settings: -fshow-warning-groups -fprefer-byte-code -fbreak-points + -fno-zap-casts warning settings: -Wpattern-namespace-specifier === :set @@ -34,6 +35,7 @@ other dynamic, non-language, flag settings: -fshow-warning-groups -fprefer-byte-code -fbreak-points + -fno-zap-casts warning settings: -Wpattern-namespace-specifier Unit ID: b-0.0.0 @@ -51,6 +53,7 @@ other dynamic, non-language, flag settings: -fshow-warning-groups -fprefer-byte-code -fbreak-points + -fno-zap-casts warning settings: -Wpattern-namespace-specifier Unit ID: c-0.0.0 @@ -68,6 +71,7 @@ other dynamic, non-language, flag settings: -fshow-warning-groups -fprefer-byte-code -fbreak-points + -fno-zap-casts warning settings: -Wpattern-namespace-specifier Unit ID: d-0.0.0 @@ -85,5 +89,6 @@ other dynamic, non-language, flag settings: -fshow-warning-groups -fprefer-byte-code -fbreak-points + -fno-zap-casts warning settings: -Wpattern-namespace-specifier View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/307ee28a3679bfdece77c6868a7da8… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/307ee28a3679bfdece77c6868a7da8… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/spj-apporv-Oct24] always add expressions in the error messages on the rhs of equations and matches
by Apoorv Ingle (＠ani) 25 Nov '25

25 Nov '25

Apoorv Ingle pushed to branch wip/spj-apporv-Oct24 at Glasgow Haskell Compiler / GHC Commits: c314ee5a by Apoorv Ingle at 2025-11-25T14:20:03-06:00 always add expressions in the error messages on the rhs of equations and matches - - - - - 2 changed files: - compiler/GHC/Tc/Gen/Head.hs - compiler/GHC/Tc/Gen/Match.hs Changes: ===================================== compiler/GHC/Tc/Gen/Head.hs ===================================== @@ -1107,12 +1107,11 @@ add_expr_ctxt e thing_inside -- Flip error ctxt into expansion mode _ -> addErrCtxt (ExprCtxt e) thing_inside - -- no op in generated code addLExprCtxt :: SrcSpan -> HsExpr GhcRn -> TcRn a -> TcRn a addLExprCtxt lspan e thing_inside | not (isGeneratedSrcSpan lspan) = setSrcSpan lspan $ add_expr_ctxt e thing_inside - | otherwise + | otherwise -- no op in generated code = thing_inside ===================================== compiler/GHC/Tc/Gen/Match.hs ===================================== @@ -42,7 +42,7 @@ import GHC.Prelude import {-# SOURCE #-} GHC.Tc.Gen.Expr( tcSyntaxOp, tcInferRho, tcInferRhoFRRNC , tcMonoLExprNC, tcMonoLExpr, tcExpr , tcCheckMonoExpr, tcCheckMonoExprNC - , tcCheckPolyExpr, tcPolyLExpr ) + , tcCheckPolyExpr, tcPolyExpr ) import GHC.Rename.Utils ( bindLocalNames ) import GHC.Tc.Errors.Types @@ -420,10 +420,15 @@ tcDoStmts mDoExpr ss@(L _ stmts) res_ty } tcBody :: LHsExpr GhcRn -> ExpRhoType -> TcM (LHsExpr GhcTc) -tcBody body res_ty - = do { traceTc "tcBody" (ppr res_ty) - ; tcPolyLExpr body res_ty - } +tcBody (L lspan e) res_ty + = L lspan <$> + do { traceTc "tcBody" (ppr res_ty) + ; setSrcSpanA lspan $ + addErrCtxt (ExprCtxt e) $ + -- We want the right hand side of a match or an equation + -- to always get printed + tcPolyExpr e res_ty + } {- ************************************************************************ View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/c314ee5aa5e9cdb0b70b0cd27a9b89b… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/c314ee5aa5e9cdb0b70b0cd27a9b89b… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/spj-apporv-Oct24] always add expressions in the error messages on the rhs of equations and matches
by Apoorv Ingle (＠ani) 25 Nov '25

25 Nov '25

Apoorv Ingle pushed to branch wip/spj-apporv-Oct24 at Glasgow Haskell Compiler / GHC Commits: 4ea84b19 by Apoorv Ingle at 2025-11-25T13:52:46-06:00 always add expressions in the error messages on the rhs of equations and matches - - - - - 2 changed files: - compiler/GHC/Tc/Gen/Head.hs - compiler/GHC/Tc/Gen/Match.hs Changes: ===================================== compiler/GHC/Tc/Gen/Head.hs ===================================== @@ -1107,12 +1107,11 @@ add_expr_ctxt e thing_inside -- Flip error ctxt into expansion mode _ -> addErrCtxt (ExprCtxt e) thing_inside - -- no op in generated code addLExprCtxt :: SrcSpan -> HsExpr GhcRn -> TcRn a -> TcRn a addLExprCtxt lspan e thing_inside | not (isGeneratedSrcSpan lspan) = setSrcSpan lspan $ add_expr_ctxt e thing_inside - | otherwise + | otherwise -- no op in generated code = thing_inside ===================================== compiler/GHC/Tc/Gen/Match.hs ===================================== @@ -42,7 +42,7 @@ import GHC.Prelude import {-# SOURCE #-} GHC.Tc.Gen.Expr( tcSyntaxOp, tcInferRho, tcInferRhoFRRNC , tcMonoLExprNC, tcMonoLExpr, tcExpr , tcCheckMonoExpr, tcCheckMonoExprNC - , tcCheckPolyExpr, tcPolyLExpr ) + , tcCheckPolyExpr, tcPolyExpr ) import GHC.Rename.Utils ( bindLocalNames ) import GHC.Tc.Errors.Types @@ -420,10 +420,15 @@ tcDoStmts mDoExpr ss@(L _ stmts) res_ty } tcBody :: LHsExpr GhcRn -> ExpRhoType -> TcM (LHsExpr GhcTc) -tcBody body res_ty - = do { traceTc "tcBody" (ppr res_ty) - ; tcPolyLExpr body res_ty - } +tcBody (L lspan e) res_ty + = L lspan <$> + do { traceTc "tcBody" (ppr res_ty) + ; setSrcSpanA lspan $ + addErrCtxt (ExprCtxt e) $ + -- We want the right hand side of a match or an equation + -- to always get printed + tcPolyExpr e res_ty) + } {- ************************************************************************ View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/4ea84b199496579abdcb9024ccbb886… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/4ea84b199496579abdcb9024ccbb886… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/spj-apporv-Oct24] reduce the number of calls to getDeepSubsumption_DataCon by passing it as an...
by Apoorv Ingle (＠ani) 25 Nov '25

25 Nov '25

Apoorv Ingle pushed to branch wip/spj-apporv-Oct24 at Glasgow Haskell Compiler / GHC Commits: f85de9bd by Apoorv Ingle at 2025-11-25T12:13:25-06:00 reduce the number of calls to getDeepSubsumption_DataCon by passing it as an argument to checkResultTy. Some error message wibbles - - - - - 3 changed files: - compiler/GHC/Tc/Gen/App.hs - compiler/GHC/Tc/Gen/Expr.hs - compiler/GHC/Tc/Gen/Head.hs Changes: ===================================== compiler/GHC/Tc/Gen/App.hs ===================================== @@ -385,7 +385,7 @@ Unify result type /before/ typechecking the args Actual: String • In the first argument of ‘Pair’, namely ‘"yes"’ -The latter is much better. That is why we call checkResultType before tcValArgs. +The latter is much better. That is why we call `checkResultTy` before tcValArgs. -} -- CAUTION: Any changes to tcApp should be reflected in tcExprSigma tcApp :: HsExpr GhcRn @@ -442,12 +442,13 @@ tcApp rn_expr exp_res_ty -- See (TCAPP1) and (TCAPP2) in -- Note [tcApp: typechecking applications] + ; ds_flag <- getDeepSubsumptionFlag_DataConHead tc_fun ; case do_ql of NoQL -> do { traceTc "tcApp:NoQL" (ppr rn_fun $$ ppr app_res_rho) -- Step 4.1: subsumption check against expected result type -- See Note [Unify with expected type before typechecking arguments] - ; res_wrap <- checkResultTy rn_expr tc_head inst_args + ; res_wrap <- checkResultTy ds_flag rn_expr tc_head inst_args app_res_rho exp_res_ty -- Step 4.2: typecheck the arguments ; tc_args <- tcValArgs NoQL (rn_fun, fun_lspan) inst_args @@ -467,7 +468,7 @@ tcApp rn_expr exp_res_ty ; app_res_rho <- liftZonkM $ zonkTcType app_res_rho -- Step 5.4: subsumption check against the expected type - ; res_wrap <- checkResultTy rn_expr tc_head inst_args + ; res_wrap <- checkResultTy ds_flag rn_expr tc_head inst_args app_res_rho exp_res_ty -- Step 5.5: wrap up ; finishApp tc_head tc_args app_res_rho res_wrap } } @@ -527,27 +528,25 @@ finishApp tc_head@(tc_fun,_) tc_args app_res_rho res_wrap -- | Connect up the inferred type of an application with the expected type. -- This is usually just a unification, but with deep subsumption there is more to do. -checkResultTy :: HsExpr GhcRn +checkResultTy :: DeepSubsumptionFlag + -> HsExpr GhcRn -> (HsExpr GhcTc, SrcSpan) -- Head -> [HsExprArg p] -- Arguments, just error messages -> TcRhoType -- Inferred type of the application; zonked to -- expose foralls, but maybe not /deeply/ instantiated -> ExpRhoType -- Expected type; this is deeply skolemised -> TcM HsWrapper -checkResultTy rn_expr (fun, _) _inst_args app_res_rho (Infer inf_res) - = do { ds_flag <- getDeepSubsumptionFlag_DataConHead fun - ; fillInferResult ds_flag (exprCtOrigin rn_expr) app_res_rho inf_res - } +checkResultTy ds_flag rn_expr (fun, _) _inst_args app_res_rho (Infer inf_res) + = fillInferResult ds_flag (exprCtOrigin rn_expr) app_res_rho inf_res -checkResultTy rn_expr (tc_fun, fun_loc) inst_args app_res_rho (Check res_ty) +checkResultTy ds_flag rn_expr (tc_fun, fun_loc) inst_args app_res_rho (Check res_ty) -- Unify with expected type from the context -- See Note [Unify with expected type before typechecking arguments] -- -- Match up app_res_rho: the result type of rn_expr -- with res_ty: the expected result type = perhaps_add_res_ty_ctxt $ - do { ds_flag <- getDeepSubsumptionFlag_DataConHead tc_fun - ; traceTc "checkResultTy {" $ + do { traceTc "checkResultTy {" $ vcat [ text "tc_fun:" <+> ppr tc_fun , text "app_res_rho:" <+> ppr app_res_rho , text "res_ty:" <+> ppr res_ty @@ -695,7 +694,7 @@ tcValArg _ pos (fun, fun_lspan) (EValArgQL { ; tc_args <- tcValArgs DoQL (rn_fun, snd tc_head) inst_args ; app_res_rho <- liftZonkM $ zonkTcType app_res_rho - ; res_wrap <- checkResultTy rn_expr tc_head inst_args + ; res_wrap <- checkResultTy ds_flag rn_expr tc_head inst_args app_res_rho (mkCheckExpType exp_arg_rho) ; finishApp tc_head tc_args app_res_rho res_wrap } ===================================== compiler/GHC/Tc/Gen/Expr.hs ===================================== @@ -21,7 +21,8 @@ module GHC.Tc.Gen.Expr tcInferRho, tcInferRhoNC, tcMonoLExpr, tcMonoLExprNC, tcInferRhoFRR, tcInferRhoFRRNC, - tcPolyLExpr, tcPolyExpr, tcExpr, tcPolyLExprSig, + tcPolyLExpr, tcPolyLExprSig, tcPolyLExprNC, + tcPolyExpr, tcExpr, tcSyntaxOp, tcSyntaxOpGen, SyntaxOpType(..), synKnownType, tcCheckId, ) where @@ -120,7 +121,7 @@ tcCheckPolyExprNC expr res_ty = tcPolyLExprNC expr (mkCheckExpType res_ty) ----------------- -- These versions take an ExpType tcPolyLExpr, tcPolyLExprNC :: LHsExpr GhcRn -> ExpSigmaType - -> TcM (LHsExpr GhcTc) + -> TcM (LHsExpr GhcTc) tcPolyLExpr (L loc expr) res_ty = addLExprCtxt (locA loc) expr $ -- Note [Error contexts in generated code] ===================================== compiler/GHC/Tc/Gen/Head.hs ===================================== @@ -269,8 +269,6 @@ splitHsApps e = go e noSrcSpan [] -- and its hard to say exactly what that is : EWrap (EExpand e) : args ) - -- look through PopErrCtxt (cf. T17594f) we do not want to lose the opportunity of calling tcEValArgQL - -- unlike HsPar, it is okay to forget about the PopErrCtxts as it does not persist over in GhcTc land go e lspan args = pure ((e, lspan), args) @@ -1087,18 +1085,16 @@ mis-match in the number of value arguments. add_expr_ctxt :: HsExpr GhcRn -> TcRn a -> TcRn a add_expr_ctxt e thing_inside = case e of - HsHole _ -> thing_inside + HsHole{} -> thing_inside -- The HsHole special case addresses situations like -- f x = _ -- when we don't want to say "In the expression: _", -- because it is mentioned in the error message itself - HsPar _ e -> add_expr_ctxt (unLoc e) thing_inside + HsPar{} -> thing_inside -- We don't want to say 'In the expression (e)', -- we just want to say 'In the expression, 'e' -- which will be handeled by the recursive call in thing_inside - -- This may be a little inefficient with nested parens exprs, eg. (((e))) - -- But it should be okay as I do not expect too many parens to be nested consecutively ExprWithTySig _ (L _ e') _ | XExpr (ExpandedThingRn o _) <- e' -> addExpansionErrCtxt o (ExprCtxt e) View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/f85de9bd132fd5e3ef8a3d4dfdf4179… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/f85de9bd132fd5e3ef8a3d4dfdf4179… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/T26425] 20 commits: Make TYPE and CONSTRAINT apart again
by Simon Peyton Jones (＠simonpj) 25 Nov '25

25 Nov '25

Simon Peyton Jones pushed to branch wip/T26425 at Glasgow Haskell Compiler / GHC Commits: a00840ea by Simon Peyton Jones at 2025-11-14T15:23:56+00:00 Make TYPE and CONSTRAINT apart again This patch finally fixes #24279. * The story started with #11715 * Then #21623 articulated a plan, which made Type and Constraint not-apart; a horrible hack but it worked. The main patch was commit 778c6adca2c995cd8a1b84394d4d5ca26b915dac Author: Simon Peyton Jones <simonpj(a)microsoft.com> Date: Wed Nov 9 10:33:22 2022 +0000 Type vs Constraint: finally nailed * #24279 reported a bug in the above big commit; this small patch fixes it commit af6932d6c068361c6ae300d52e72fbe13f8e1f18 Author: Simon Peyton Jones <simon.peytonjones(a)gmail.com> Date: Mon Jan 8 10:49:49 2024 +0000 Make TYPE and CONSTRAINT not-apart Issue #24279 showed up a bug in the logic in GHC.Core.Unify.unify_ty which is supposed to make TYPE and CONSTRAINT be not-apart. * Then !10479 implemented "unary classes". * That change in turn allows us to make Type and Constraint apart again, cleaning up the compiler and allowing a little bit more expressiveness. It fixes the original hope in #24279, namely that `Type` and `Constraint` should be distinct throughout. - - - - - c0a1e574 by Georgios Karachalias at 2025-11-15T05:14:31-05:00 Report all missing modules with -M We now report all missing modules at once in GHC.Driver.Makefile.processDeps, as opposed to only reporting a single missing module. Fixes #26551. - - - - - c9fa3449 by Sylvain Henry at 2025-11-15T05:15:26-05:00 JS: fix array index for registers We used to store R32 in h$regs[-1]. While it's correct in JavaScript, fix this to store R32 in h$regs[0] instead. - - - - - 9e469909 by Sylvain Henry at 2025-11-15T05:15:26-05:00 JS: support more than 128 registers (#26558) The JS backend only supported 128 registers (JS variables/array slots used to pass function arguments). It failed in T26537 when 129 registers were required. This commit adds support for more than 128 registers: it is now limited to maxBound :: Int (compiler's Int). If we ever go above this threshold the compiler now panics with a more descriptive message. A few built-in JS functions were assuming 128 registers and have been rewritten to use loops. Note that loops are only used for "high" registers that are stored in an array: the 31 "low" registers are still handled with JS global variables and with explicit switch-cases to maintain good performance in the most common cases (i.e. few registers used). Adjusting the number of low registers is now easy: just one constant to adjust (GHC.StgToJS.Regs.lowRegsCount). No new test added: T26537 is used as a regression test instead. - - - - - 0a64a78b by Sven Tennie at 2025-11-15T20:31:10-05:00 AArch64: Simplify CmmAssign and CmmStore The special handling for floats was fake: The general case is always used. So, the additional code path isn't needed (and only adds complexity for the reader.) - - - - - 15b311be by sheaf at 2025-11-15T20:32:02-05:00 SimpleOpt: refactor & push coercions into lambdas This commit improves the simple optimiser (in GHC.Core.SimpleOpt) in a couple of ways: - The logic to push coercion lambdas is shored up. The function 'pushCoercionIntoLambda' used to be called in 'finish_app', but this meant we could not continue to optimise the program after performing this transformation. Now, we call 'pushCoercionIntoLambda' as part of 'simple_app'. Doing so can be important when dealing with unlifted newtypes, as explained in Note [Desugaring unlifted newtypes]. - The code is re-structured to avoid duplication and out-of-sync code paths. Now, 'simple_opt_expr' defers to 'simple_app' for the 'App', 'Var', 'Cast' and 'Lam' cases. This means all the logic for those is centralised in a single place (e.g. the 'go_lam' helper function). To do this, the general structure is brought a bit closer to the full-blown simplifier, with a notion of 'continuation' (see 'SimpleContItem'). This commit also modifies GHC.Core.Opt.Arity.pushCoercionIntoLambda to apply a substitution (a slight generalisation of its existing implementation). - - - - - b33284c7 by sheaf at 2025-11-15T20:32:02-05:00 Improve typechecking of data constructors This commit changes the way in which we perform typecheck data constructors, in particular how we make multiplicities line up. Now, impedance matching occurs as part of the existing subsumption machinery. See the revamped Note [Typechecking data constructors] in GHC.Tc.Gen.App, as well as Note [Polymorphisation of linear fields] in GHC.Core.Multiplicity. This allows us to get rid of a fair amount of hacky code that was added with the introduction of LinearTypes; in particular the logic of GHC.Tc.Gen.Head.tcInferDataCon. ------------------------- Metric Decrease: T10421 T14766 T15164 T15703 T19695 T5642 T9630 WWRec ------------------------- - - - - - b6faf5d0 by sheaf at 2025-11-15T20:32:02-05:00 Handle unsaturated rep-poly newtypes This commit allows GHC to handle unsaturated occurrences of unlifted newtype constructors. The plan is detailed in Note [Eta-expanding rep-poly unlifted newtypes] in GHC.Tc.Utils.Concrete: for unsaturated unlifted newtypes, we perform the appropriate representation-polymorphism check in tcInstFun. - - - - - 682bf979 by Mike Pilgrem at 2025-11-16T16:44:14+00:00 Fix #26293 Valid stack.yaml for hadrian - - - - - acc70c3a by Simon Peyton Jones at 2025-11-18T16:21:20-05:00 Fix a bug in defaulting Addresses #26582 Defaulting was doing some unification but then failing to iterate. Silly. I discovered that the main solver was unnecessarily iterating even if there was a unification for an /outer/ unification variable, so I fixed that too. - - - - - c12fa73e by Simon Peyton Jones at 2025-11-19T02:55:01-05:00 Make PmLit be in Ord, and use it in Map This MR addresses #26514, by changing from data PmAltConSet = PACS !(UniqDSet ConLike) ![PmLit] to data PmAltConSet = PACS !(UniqDSet ConLike) !(Map PmLit PmLit) This matters when doing pattern-match overlap checking, when there is a very large set of patterns. For most programs it makes no difference at all. For the N=5000 case of the repro case in #26514, compiler mutator time (with `-fno-code`) goes from 1.9s to 0.43s. All for the price for an Ord instance for PmLit - - - - - 41b84f40 by sheaf at 2025-11-19T02:55:52-05:00 Add passing tests for #26311 and #26072 This commit adds two tests cases that now pass since landing the changes to typechecking of data constructors in b33284c7. Fixes #26072 #26311 - - - - - 1faa758a by sheaf at 2025-11-19T02:55:52-05:00 mkCast: weaken bad cast warning for multiplicity This commit weakens the warning message emitted when constructing a bad cast in mkCast to ignore multiplicity. Justification: since b33284c7, GHC uses sub-multiplicity coercions to typecheck data constructors. The coercion optimiser is free to discard these coercions, both for performance reasons, and because GHC's Core simplifier does not (yet) preserve linearity. We thus weaken 'mkCast' to use 'eqTypeIgnoringMultiplicity' instead of 'eqType', to avoid getting many spurious warnings about mismatched multiplicities. - - - - - 55eab80d by Sylvain Henry at 2025-11-20T17:33:13-05:00 Build external interpreter program on demand (#24731) This patch teaches GHC how to build the external interpreter program when it is missing. As long as we have the `ghci` library, doing this is trivial so most of this patch is refactoring for doing it sanely. - - - - - 08bbc028 by Rodrigo Mesquita at 2025-11-20T17:33:54-05:00 Add tests for #23973 and #26565 These were fixed by 4af4f0f070f83f948e49ad5d7835fd91b8d3f0e6 in !10417 - - - - - 6b42232c by sheaf at 2025-11-20T17:34:35-05:00 Mark T26410_ffi as fragile on Windows As seen in #26595, this test intermittently fails on Windows. This commit marks it as fragile, until we get around to fixing it. - - - - - b7b7c049 by Andrew Lelechenko at 2025-11-21T21:04:01+00:00 Add nubOrd / nubOrdBy to Data.List and Data.List.NonEmpty As per https://github.com/haskell/core-libraries-committee/issues/336 - - - - - 352d5462 by Marc Scholten at 2025-11-22T10:33:03-05:00 Fix haddock test runner to handle UTF-8 output xhtml 3000.4.0.0 now produces UTF-8 output instead of escaping non-ASCII characters. When using --test-accept it previously wrote files in the wrong encoding because they have not been decoded properly when reading the files. - - - - - 48a3ed57 by Simon Peyton Jones at 2025-11-25T15:33:54+00:00 Add a fast-path for args=[] to occAnalApp In the common case of having not arguments, occAnalApp was doing redundant work. - - - - - 951e5ed9 by Simon Peyton Jones at 2025-11-25T15:33:54+00:00 Fix a performance hole in the occurrence analyser As #26425 showed, the clever stuff in Note [Occurrence analysis for join points] does a lot of duplication of usage details. This patch improved matters with a little fancy footwork. It is described in the new (W4) of the same Note. Compile-time allocations go down slightly. Here are the changes of +/- 0.5% or more: T13253(normal) 329,369,244 326,395,544 -0.9% T13253-spj(normal) 66,410,496 66,095,864 -0.5% T15630(normal) 129,797,200 128,663,136 -0.9% T15630a(normal) 129,212,408 128,027,560 -0.9% T16577(normal) 6,756,706,896 6,723,028,512 -0.5% T18282(normal) 128,462,070 125,808,584 -2.1% GOOD T18698a(normal) 208,418,305 202,037,336 -3.1% GOOD T18730(optasm) 136,981,756 136,208,136 -0.6% T18923(normal) 58,103,088 57,745,840 -0.6% T19695(normal) 1,386,306,272 1,365,609,416 -1.5% T26425(normal) 3,344,402,957 2,457,811,664 -26.5% GOOD T6048(optasm) 79,763,816 79,212,760 -0.7% T9020(optasm) 225,278,408 223,682,440 -0.7% T9961(normal) 303,810,717 300,729,168 -1.0% GOOD geo. mean -0.5% minimum -26.5% maximum +0.4% Metric Decrease: T18282 T18698a T26425 T9961 - - - - - 147 changed files: - compiler/GHC.hs - compiler/GHC/Builtin/Types/Prim.hs - compiler/GHC/CmmToAsm/AArch64/CodeGen.hs - compiler/GHC/Core.hs - compiler/GHC/Core/Coercion.hs - compiler/GHC/Core/Coercion/Opt.hs - compiler/GHC/Core/DataCon.hs - compiler/GHC/Core/Lint.hs - compiler/GHC/Core/Multiplicity.hs - compiler/GHC/Core/Opt/Arity.hs - compiler/GHC/Core/Opt/OccurAnal.hs - compiler/GHC/Core/Opt/Simplify/Iteration.hs - compiler/GHC/Core/RoughMap.hs - compiler/GHC/Core/SimpleOpt.hs - compiler/GHC/Core/TyCo/FVs.hs - compiler/GHC/Core/TyCo/Rep.hs - compiler/GHC/Core/Type.hs - compiler/GHC/Core/Unify.hs - compiler/GHC/Core/Utils.hs - compiler/GHC/Driver/Config/Core/Lint.hs - + compiler/GHC/Driver/Config/Interpreter.hs - compiler/GHC/Driver/Config/Linker.hs - compiler/GHC/Driver/DynFlags.hs - compiler/GHC/Driver/MakeFile.hs - compiler/GHC/Driver/Pipeline.hs - compiler/GHC/Driver/Pipeline/Execute.hs - compiler/GHC/Hs/Expr.hs - compiler/GHC/Hs/Syn/Type.hs - compiler/GHC/Hs/Utils.hs - compiler/GHC/HsToCore/Binds.hs - compiler/GHC/HsToCore/Expr.hs - compiler/GHC/HsToCore/Match.hs - compiler/GHC/HsToCore/Pmc/Solver/Types.hs - compiler/GHC/HsToCore/Utils.hs - compiler/GHC/Iface/Ext/Ast.hs - compiler/GHC/Iface/Type.hs - compiler/GHC/Linker/Config.hs - compiler/GHC/Linker/Dynamic.hs - + compiler/GHC/Linker/Executable.hs - − compiler/GHC/Linker/ExtraObj.hs - compiler/GHC/Linker/Loader.hs - compiler/GHC/Linker/MacOS.hs - compiler/GHC/Linker/Static.hs - compiler/GHC/Linker/Windows.hs - + compiler/GHC/Runtime/Interpreter/C.hs - + compiler/GHC/Runtime/Interpreter/Init.hs - compiler/GHC/StgToJS/Apply.hs - compiler/GHC/StgToJS/Expr.hs - compiler/GHC/StgToJS/Regs.hs - compiler/GHC/StgToJS/Rts/Rts.hs - compiler/GHC/StgToJS/Rts/Types.hs - compiler/GHC/SysTools/Tasks.hs - compiler/GHC/Tc/Gen/App.hs - compiler/GHC/Tc/Gen/Expr.hs - compiler/GHC/Tc/Gen/Expr.hs-boot - compiler/GHC/Tc/Gen/Head.hs - compiler/GHC/Tc/Gen/Match.hs - compiler/GHC/Tc/Gen/Pat.hs - compiler/GHC/Tc/Instance/Class.hs - compiler/GHC/Tc/Solver/Default.hs - compiler/GHC/Tc/Solver/Monad.hs - compiler/GHC/Tc/Solver/Solve.hs - compiler/GHC/Tc/Types/Evidence.hs - compiler/GHC/Tc/Types/Origin.hs - compiler/GHC/Tc/Utils/Concrete.hs - compiler/GHC/Tc/Utils/TcType.hs - compiler/GHC/Tc/Utils/Unify.hs - compiler/GHC/Tc/Zonk/Type.hs - compiler/GHC/Types/Basic.hs - compiler/GHC/Types/Id/Make.hs - compiler/GHC/Types/SourceText.hs - compiler/GHC/Types/Unique/FM.hs - compiler/GHC/Types/Unique/Set.hs - compiler/GHC/Types/Var/Env.hs - compiler/ghc.cabal.in - docs/users_guide/9.16.1-notes.rst - hadrian/stack.yaml - hadrian/stack.yaml.lock - libraries/base/base.cabal.in - libraries/base/changelog.md - libraries/base/src/Data/List.hs - libraries/base/src/Data/List/NonEmpty.hs - + libraries/base/src/Data/List/NubOrdSet.hs - libraries/ghc-internal/src/GHC/Internal/Data/OldList.hs - + testsuite/tests/bytecode/T23973.hs - + testsuite/tests/bytecode/T23973.script - + testsuite/tests/bytecode/T23973.stdout - + testsuite/tests/bytecode/T26565.hs - + testsuite/tests/bytecode/T26565.script - + testsuite/tests/bytecode/T26565.stdout - testsuite/tests/bytecode/all.T - testsuite/tests/codeGen/should_run/all.T - testsuite/tests/diagnostic-codes/codes.stdout - testsuite/tests/driver/Makefile - + testsuite/tests/driver/T24731.hs - + testsuite/tests/driver/T26551.hs - + testsuite/tests/driver/T26551.stderr - testsuite/tests/driver/all.T - testsuite/tests/ghci/scripts/T8959b.stderr - testsuite/tests/ghci/scripts/ghci051.stderr - testsuite/tests/indexed-types/should_compile/T12538.stderr - testsuite/tests/indexed-types/should_fail/T21092.hs - − testsuite/tests/indexed-types/should_fail/T21092.stderr - testsuite/tests/indexed-types/should_fail/all.T - testsuite/tests/interface-stability/base-exports.stdout - testsuite/tests/interface-stability/base-exports.stdout-javascript-unknown-ghcjs - testsuite/tests/interface-stability/base-exports.stdout-mingw32 - testsuite/tests/interface-stability/base-exports.stdout-ws-32 - + testsuite/tests/linear/should_compile/LinearEtaExpansions.hs - testsuite/tests/linear/should_compile/all.T - testsuite/tests/linear/should_fail/TypeClass.hs - testsuite/tests/linear/should_fail/TypeClass.stderr - testsuite/tests/linear/should_run/LinearGhci.stdout - + testsuite/tests/linear/should_run/T26311.hs - + testsuite/tests/linear/should_run/T26311.stdout - testsuite/tests/linear/should_run/all.T - testsuite/tests/numeric/should_compile/T16402.stderr-ws-64 - testsuite/tests/parser/should_compile/DumpTypecheckedAst.stderr - testsuite/tests/perf/compiler/all.T - testsuite/tests/pmcheck/should_compile/pmcOrPats.stderr - testsuite/tests/rep-poly/RepPolyCase1.stderr - − testsuite/tests/rep-poly/RepPolyCase2.stderr - testsuite/tests/rep-poly/RepPolyRule3.stderr - testsuite/tests/rep-poly/RepPolyTuple4.stderr - testsuite/tests/rep-poly/T13233.stderr - − testsuite/tests/rep-poly/T17021.stderr - testsuite/tests/rep-poly/T20363b.stderr - − testsuite/tests/rep-poly/T21650_a.stderr - − testsuite/tests/rep-poly/T21650_b.stderr - + testsuite/tests/rep-poly/T26072.hs - + testsuite/tests/rep-poly/T26072b.hs - testsuite/tests/rep-poly/UnliftedNewtypesLevityBinder.stderr - testsuite/tests/rep-poly/all.T - testsuite/tests/simd/should_run/all.T - + testsuite/tests/typecheck/should_compile/T26582.hs - testsuite/tests/typecheck/should_compile/all.T - testsuite/tests/typecheck/should_fail/T15883e.stderr - testsuite/tests/typecheck/should_fail/T2414.stderr - testsuite/tests/typecheck/should_fail/T24279.hs - − testsuite/tests/typecheck/should_fail/T24279.stderr - testsuite/tests/typecheck/should_fail/T2534.stderr - testsuite/tests/typecheck/should_fail/T7264.stderr - testsuite/tests/typecheck/should_fail/all.T - utils/haddock/haddock-test/src/Test/Haddock.hs - utils/haddock/hypsrc-test/ref/src/Classes.html - utils/haddock/hypsrc-test/ref/src/Quasiquoter.html - utils/iserv/iserv.cabal.in The diff was not included because it is too large. View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/6b5e7254ad0f0d3bd68b1972934e89… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/6b5e7254ad0f0d3bd68b1972934e89… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/amg/castz] Fumble around trying to fix optimisation bugs
by Adam Gundry (＠adamgundry) 25 Nov '25

25 Nov '25

Adam Gundry pushed to branch wip/amg/castz at Glasgow Haskell Compiler / GHC Commits: 307ee28a by Adam Gundry at 2025-11-25T14:26:41+00:00 Fumble around trying to fix optimisation bugs - - - - - 3 changed files: - compiler/GHC/Core/Opt/Arity.hs - compiler/GHC/Core/Opt/Simplify/Iteration.hs - compiler/GHC/Core/SimpleOpt.hs Changes: ===================================== compiler/GHC/Core/Opt/Arity.hs ===================================== @@ -2246,13 +2246,13 @@ etaInfoApp in_scope expr eis -- Beta-reduction if possible, pushing any intervening casts past -- the argument. See Note [The EtaInfo mechanism] go subst (Lam v e) (EI (b:bs) mco) - | Just (arg,mco') <- pushCoArg (exprType (Lam v e)) mco (varToCoreExpr b) + | Just (arg,_, mco') <- pushCoArg (exprType (Lam v e)) mco (varToCoreExpr b) = go (Core.extendSubst subst v arg) e (EI bs mco') -- Stop pushing down; just wrap the expression up -- See Note [Check for reflexive casts in eta expansion] go subst e (EI bs mco) = Core.substExprSC subst e - `mkCastCo` checkReflexiveCastCo (exprType e) mco -- TODO check type + `mkCastCo` checkReflexiveCastCo (exprType e) mco `mkVarApps` bs -------------- @@ -2359,12 +2359,12 @@ mkEtaWW orig_oss ppr_orig_expr in_scope orig_ty -- with an explicit lambda having a non-function type mkEtaForAllMCo :: ForAllTyBinder -> Type -> CastCoercion -> CastCoercion -mkEtaForAllMCo (Bndr tcv vis) ty mco +mkEtaForAllMCo bdnr@(Bndr tcv vis) ty mco = case mco of ReflCastCo | vis == coreTyLamForAllTyFlag -> ReflCastCo | otherwise -> mk_fco (mkRepReflCo ty) CCoercion co -> mk_fco co - ZCoercion _ty2 cos -> ZCoercion ty cos -- TODO: is ty right? + ZCoercion tyR cos -> ZCoercion (mkForAllTy bdnr tyR) cos where mk_fco co = CCoercion (mkForAllCo tcv vis coreTyLamForAllTyFlag MRefl co) -- coreTyLamForAllTyFlag: See Note [The EtaInfo mechanism], particularly @@ -2871,17 +2871,17 @@ Here we implement the "push rules" from FC papers: by pushing the coercion into the arguments -} -pushCoArgs :: Type -> CastCoercion -> [CoreArg] -> Maybe ([CoreArg], CastCoercion) -pushCoArgs _ co [] = return ([], co) +pushCoArgs :: Type -> CastCoercion -> [CoreArg] -> Maybe ([CoreArg], Type, CastCoercion) +pushCoArgs fun_ty co [] = return ([], fun_ty, co) pushCoArgs fun_ty co (arg:args) = do - { (arg', m_co1) <- pushCoArg fun_ty co arg + { (arg', ty, m_co1) <- pushCoArg fun_ty co arg ; if isReflCastCo m_co1 - then return (arg':args, ReflCastCo) - else do { (args', m_co2) <- pushCoArgs (funResultTy fun_ty) m_co1 args -- TODO check type - ; return (arg':args', m_co2) } + then return (arg':args, ty, ReflCastCo) + else do { (args', ty', m_co2) <- pushCoArgs ty m_co1 args + ; return (arg':args', ty', m_co2) } } -pushCoArg :: Type -> CastCoercion -> CoreArg -> Maybe (CoreArg, CastCoercion) +pushCoArg :: Type -> CastCoercion -> CoreArg -> Maybe (CoreArg, Type, CastCoercion) -- We have (fun |> co) arg, and we want to transform it to -- (fun arg) |> co -- This may fail, e.g. if (fun :: N) where N is a newtype @@ -2889,22 +2889,22 @@ pushCoArg :: Type -> CastCoercion -> CoreArg -> Maybe (CoreArg, CastCoercion) -- 'co' is always Representational pushCoArg fun_ty co arg | Type ty <- arg - = do { (ty', m_co') <- pushCastCoTyArg co ty - ; return (Type ty', m_co') } + = do { (ty', ty, m_co') <- pushCastCoTyArg co ty + ; return (Type ty', ty, m_co') } | otherwise - = do { (arg_mco, m_co') <- pushCastCoValArg fun_ty co - ; let arg_mco' = checkReflexiveCastCo (funArgTy fun_ty) arg_mco + = do { (arg_ty, arg_mco, res_ty, m_co') <- pushCastCoValArg fun_ty co + ; let arg_mco' = checkReflexiveCastCo arg_ty arg_mco -- checkReflexiveCastCo: see Note [Check for reflexive casts in eta expansion] -- The coercion is very often (arg_co -> res_co), but without -- the argument coercion actually being ReflCo - ; return (arg `mkCastCo` arg_mco', m_co') } + ; return (arg `mkCastCo` arg_mco', res_ty, m_co') } -pushCastCoTyArg :: CastCoercion -> Type -> Maybe (Type, CastCoercion) +pushCastCoTyArg :: CastCoercion -> Type -> Maybe (Type, Type, CastCoercion) pushCastCoTyArg (CCoercion co) ty = pushCoTyArg co ty -pushCastCoTyArg ReflCastCo ty = Just (ty, ReflCastCo) +pushCastCoTyArg ReflCastCo ty = Just (ty, error "TODO: asdasdad", ReflCastCo) pushCastCoTyArg (ZCoercion _fun_ty _cos) _ty = Nothing -- TODO do better -pushCoTyArg :: CoercionR -> Type -> Maybe (Type, CastCoercion) +pushCoTyArg :: CoercionR -> Type -> Maybe (Type, Type, CastCoercion) -- We have (fun |> co) @ty -- Push the coercion through to return -- (fun @ty') |> co' @@ -2916,11 +2916,11 @@ pushCoTyArg co ty -- -- = Just (ty, Nothing) | isReflCo co - = Just (ty, ReflCastCo) + = Just (ty, coercionLKind co2, ReflCastCo) | isForAllTy_ty tyL = assertPpr (isForAllTy_ty tyR) (ppr co $$ ppr ty) $ - Just (ty `mkCastTy` co1, CCoercion co2) + Just (ty `mkCastTy` co1, coercionLKind co2, CCoercion co2) | otherwise = Nothing @@ -2940,14 +2940,14 @@ pushCoTyArg co ty -- co2 :: ty1[ (ty|>co1)/a1 ] ~R ty2[ ty/a2 ] -- Arg of mkInstCo is always nominal, hence Nominal -pushCastCoValArg :: Type -> CastCoercion -> Maybe (CastCoercion, CastCoercion) -pushCastCoValArg _ ReflCastCo = Just (ReflCastCo, ReflCastCo) +pushCastCoValArg :: Type -> CastCoercion -> Maybe (Type, CastCoercion, Type, CastCoercion) +pushCastCoValArg tyL ReflCastCo = Just (funArgTy tyL, ReflCastCo, funResultTy tyL, ReflCastCo) pushCastCoValArg _ (CCoercion co) = pushCoValArg co pushCastCoValArg tyL (ZCoercion tyR cos) | isFunTy tyL -- TODO: do we need to check this or can we assume it? , isFunTy tyR , typeHasFixedRuntimeRep new_arg_ty - = Just (ZCoercion new_arg_ty cos, ZCoercion (funResultTy tyR) cos) + = Just (funArgTy tyL, ZCoercion new_arg_ty cos, funResultTy tyL, ZCoercion (funResultTy tyR) cos) | otherwise = Nothing where new_arg_ty = funArgTy tyR @@ -2963,7 +2963,7 @@ pushCastCoValArg tyL (ZCoercion tyR cos) -- If the LHS is well-typed, then so is the RHS. In particular, the argument -- @arg |> co_arg@ is guaranteed to have a fixed 'RuntimeRep', in the sense of -- Note [Fixed RuntimeRep] in GHC.Tc.Utils.Concrete. -pushCoValArg :: CoercionR -> Maybe (CastCoercion, CastCoercion) +pushCoValArg :: CoercionR -> Maybe (Type, CastCoercion, Type, CastCoercion) pushCoValArg co -- The following is inefficient - don't do `eqType` here, the coercion -- optimizer will take care of it. See #14737. @@ -2971,7 +2971,7 @@ pushCoValArg co -- -- = Just (mkRepReflCo arg, Nothing) | isReflCo co - = Just (ReflCastCo, ReflCastCo) + = Just (old_arg_ty, ReflCastCo, funResultTy tyL, ReflCastCo) | isFunTy tyL , (_, co1, co2) <- decomposeFunCo co @@ -2990,7 +2990,7 @@ pushCoValArg co (vcat [ text "co:" <+> ppr co , text "old_arg_ty:" <+> ppr old_arg_ty , text "new_arg_ty:" <+> ppr new_arg_ty ]) $ - Just (coToCastCo (mkSymCo co1), coToCastCo co2) + Just (old_arg_ty, coToCastCo (mkSymCo co1), funResultTy tyL, coToCastCo co2) -- Critically, coToCastCo to checks for ReflCo; the whole coercion may not -- be reflexive, but either of its components might be -- We could use isReflexiveCo, but it's not clear if the benefit ===================================== compiler/GHC/Core/Opt/Simplify/Iteration.hs ===================================== @@ -1715,10 +1715,10 @@ simplCast env body co0 cont0 -- False: (mkTransCo co1 co2) is not fully optimised -- See Note [Avoid re-simplifying coercions] - addCoerce tyL co co_is_opt (ApplyToTy { sc_arg_ty = arg_ty, sc_hole_ty = hole_ty, sc_cont = tail }) - | Just (arg_ty', m_co') <- pushCastCoTyArg co arg_ty + addCoerce tyL co co_is_opt (ApplyToTy { sc_arg_ty = arg_ty, sc_cont = tail }) + | Just (arg_ty', res_ty, m_co') <- pushCastCoTyArg co arg_ty = {-#SCC "addCoerce-pushCoTyArg" #-} - do { tail' <- addCoerceM hole_ty m_co' co_is_opt tail -- TODO is hole_ty right? + do { tail' <- addCoerceM res_ty m_co' co_is_opt tail ; return (ApplyToTy { sc_arg_ty = arg_ty' , sc_cont = tail' , sc_hole_ty = tyL }) } @@ -1734,9 +1734,9 @@ simplCast env body co0 cont0 | not co_is_opt -- pushCoValArg duplicates the coercion, so optimise first = addCoerce tyL (optOutCastCoercion (zapSubstEnv env) co co_is_opt) True cont - | Just (m_co1, m_co2) <- pushCastCoValArg fun_ty co -- TODO check fun_ty + | Just (_, m_co1, res_ty, m_co2) <- pushCastCoValArg tyL co = {-#SCC "addCoerce-pushCoValArg" #-} - do { tail' <- addCoerceM (funResultTy fun_ty) m_co2 co_is_opt tail -- TODO check funResultTy fun_ty + do { tail' <- addCoerceM res_ty m_co2 co_is_opt tail ; if isReflCastCo m_co1 then return (cont { sc_cont = tail' , sc_hole_ty = tyL }) ; ===================================== compiler/GHC/Core/SimpleOpt.hs ===================================== @@ -1376,7 +1376,7 @@ exprIsConApp_maybe ise@(ISE in_scope id_unf) expr | not (tickishIsCode t) = go subst floats expr cont go subst floats (Cast expr co1) (CC args m_co2) -- TODO: is the subst_ty below needed? - | Just (args', m_co1') <- pushCoArgs (subst_ty subst (exprType expr)) (subst_cast_co subst co1) args + | Just (args', _, m_co1') <- pushCoArgs (subst_ty subst (exprType expr)) (subst_cast_co subst co1) args -- See Note [Push coercions in exprIsConApp_maybe] = go subst floats expr (CC args' (m_co1' `mkTransCastCo` m_co2)) View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/307ee28a3679bfdece77c6868a7da89… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/307ee28a3679bfdece77c6868a7da89… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/ghc-9.12-fix-wasm] docs: add wasm related items in 9.12.3 notes
by Cheng Shao (＠TerrorJack) 25 Nov '25

25 Nov '25

Cheng Shao pushed to branch wip/ghc-9.12-fix-wasm at Glasgow Haskell Compiler / GHC Commits: 3cd27895 by Cheng Shao at 2025-11-25T14:30:05+01:00 docs: add wasm related items in 9.12.3 notes - - - - - 1 changed file: - docs/users_guide/9.12.3-notes.rst Changes: ===================================== docs/users_guide/9.12.3-notes.rst ===================================== @@ -31,6 +31,7 @@ Compiler - Fixed GHC settings to always unescape escaped spaces (:ghc-ticket:`25204`) - Fixed issue with HasCallStack constraint caching (:ghc-ticket:`25529`) - Fixed archive member size writing logic in ``GHC.SysTools.Ar`` (:ghc-ticket:`26120`, :ghc-ticket:`22586`) +- Added support for synchronous JSFFI exports for the wasm backend (:ref:`wasm-jsffi`) Runtime System ~~~~~~~~~~~~~~ @@ -46,6 +47,7 @@ Runtime System - Improved documentation of SLIDE and PACK bytecode instructions - Eliminated redundant ``SLIDE x 0`` bytecode instructions - Fixed compile issues on powerpc64 ELF v1 +- Various fixes of wasm-specific logic in the RTS Code Generation ~~~~~~~~~~~~~~~ @@ -142,5 +144,3 @@ for further change information. libraries/xhtml/xhtml.cabal: Dependency of ``haddock`` executable utils/haddock/haddock-api/haddock-api.cabal: Dependency of ``haddock`` executable utils/haddock/haddock-library/haddock-library.cabal: Dependency of ``haddock`` executable - - View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/3cd27895944c5f76ce606008476114e… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/3cd27895944c5f76ce606008476114e… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/backports-9.14] rts: fix eager black holes: record mutated closure and fix assertion
by Zubin (＠wz1000) 25 Nov '25

25 Nov '25

Zubin pushed to branch wip/backports-9.14 at Glasgow Haskell Compiler / GHC Commits: bcc68c56 by Luite Stegeman at 2025-11-25T18:23:46+05:30 rts: fix eager black holes: record mutated closure and fix assertion This fixes two problems with handling eager black holes, introduced by a1de535f762bc23d4cf23a5b1853591dda12cdc9. - the closure mutation must be recorded even for eager black holes, since the mutator has mutated it before calling threadPaused - The assertion that an unmarked eager black hole must be owned by the TSO calling threadPaused is incorrect, since multiple threads can race to claim the black hole. fixes #26495 (cherry picked from commit 3ba3d9f9db784c903ebe8fd617447ce62d30b7d3) - - - - - 1 changed file: - rts/ThreadPaused.c Changes: ===================================== rts/ThreadPaused.c ===================================== @@ -321,11 +321,12 @@ threadPaused(Capability *cap, StgTSO *tso) if(frame_info == &stg_bh_upd_frame_info) { // eager black hole: we do nothing - // it should be a black hole that we own + // it should be a black hole (but we may not own it, as another + // thread could have raced us to claim it) ASSERT(bh_info == &stg_BLACKHOLE_info || bh_info == &__stg_EAGER_BLACKHOLE_info || bh_info == &stg_CAF_BLACKHOLE_info); - ASSERT(blackHoleOwner(bh) == tso || blackHoleOwner(bh) == NULL); + } else { // lazy black hole @@ -368,13 +369,15 @@ threadPaused(Capability *cap, StgTSO *tso) // The payload of the BLACKHOLE points to the TSO RELEASE_STORE(&((StgInd *)bh)->indirectee, (StgClosure *)tso); SET_INFO_RELEASE(bh,&stg_BLACKHOLE_info); + } - // .. and we need a write barrier, since we just mutated the closure: - recordClosureMutated(cap,bh); + // We need a write barrier, since the closure was mutated (by + // threadPaused for lazy black holes, or the mutator for eager + // black holes). + recordClosureMutated(cap,bh); - // We pretend that bh has just been created. - LDV_RECORD_CREATE(bh); - } + // We pretend that bh has just been created. + LDV_RECORD_CREATE(bh); frame = (StgClosure *) ((StgUpdateFrame *)frame + 1); if (prev_was_update_frame) { View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/bcc68c5623d5b51c5daabbd7445edfa… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/bcc68c5623d5b51c5daabbd7445edfa… You're receiving this email because of your account on gitlab.haskell.org.

1 0