- ghc-commits - Haskell.org

[Git][ghc/ghc][wip/T26110] compiler: Fix seq'ing of typecheck output for timing
by Ben Gamari (＠bgamari) 12 Jun '25

12 Jun '25

Ben Gamari pushed to branch wip/T26110 at Glasgow Haskell Compiler / GHC Commits: bb3b2fa7 by Ben Gamari at 2025-06-12T13:38:55-04:00 compiler: Fix seq'ing of typecheck output for timing As noted in #26110, previously the `withTiming` call characterising typechecking would fail to force anything at all. This meant that laziness could easily compromise the accuracy of timings. We now try to force a reasonable subset of `TcGblEnv`. I do *not* force the entire structure as many of its constituents have no easy way to reduce them to normal form. Fixes #26110. - - - - - 2 changed files: - compiler/GHC/Tc/Module.hs - compiler/GHC/Tc/Types.hs Changes: ===================================== compiler/GHC/Tc/Module.hs ===================================== @@ -211,7 +211,7 @@ tcRnModule hsc_env mod_sum save_rn_syntax | RealSrcSpan real_loc _ <- loc = withTiming logger (text "Renamer/typechecker"<+>brackets (ppr this_mod)) - (const ()) $ + (maybe () seqTcGblEnv . snd) $ initTc hsc_env hsc_src save_rn_syntax this_mod real_loc $ withTcPlugins hsc_env $ withDefaultingPlugins hsc_env $ ===================================== compiler/GHC/Tc/Types.hs ===================================== @@ -30,7 +30,7 @@ module GHC.Tc.Types( -- The environment types Env(..), - TcGblEnv(..), TcLclEnv(..), modifyLclCtxt, TcLclCtxt(..), + TcGblEnv(..), seqTcGblEnv, TcLclEnv(..), modifyLclCtxt, TcLclCtxt(..), setLclEnvTcLevel, getLclEnvTcLevel, setLclEnvLoc, getLclEnvLoc, lclEnvInGeneratedCode, IfGblEnv(..), IfLclEnv(..), @@ -171,10 +171,11 @@ import GHC.Unit.Module.Deps import GHC.Unit.Module.ModDetails import GHC.Utils.Error -import GHC.Utils.Outputable import GHC.Utils.Fingerprint -import GHC.Utils.Panic import GHC.Utils.Logger +import GHC.Utils.Misc ( seqList ) +import GHC.Utils.Outputable +import GHC.Utils.Panic import GHC.Builtin.Names ( isUnboundName ) @@ -697,6 +698,19 @@ data TcGblEnv -- ^ See Note [Generating fresh names for FFI wrappers] } +-- | Force the final result bits of a 'TcGblEnv'. This is used +-- to +seqTcGblEnv :: TcGblEnv -> () +seqTcGblEnv env = + tcg_type_env env `seq` + tcg_ann_env env `seq` + maybe () (`seqList` ()) (tcg_rn_exports env) `seq` + seqList (tcg_import_decls env) `seq` + seqList (tcg_merged env) `seq` + maybe () (`seq` ()) (tcg_rn_decls env) `seq` + tcg_dus env `seq` + () + -- NB: topModIdentity, not topModSemantic! -- Definition sites of orphan identities will be identity modules, not semantic -- modules. View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/bb3b2fa7af0cd02a96c222375565457… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/bb3b2fa7af0cd02a96c222375565457… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc] Pushed new branch wip/T26110
by Ben Gamari (＠bgamari) 12 Jun '25

12 Jun '25

Ben Gamari pushed new branch wip/T26110 at Glasgow Haskell Compiler / GHC -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/tree/wip/T26110 You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc] Pushed new branch wip/T130103
by Ben Gamari (＠bgamari) 12 Jun '25

12 Jun '25

Ben Gamari pushed new branch wip/T130103 at Glasgow Haskell Compiler / GHC -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/tree/wip/T130103 You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc] Pushed new branch wip/ghc-9.14
by Ben Gamari (＠bgamari) 12 Jun '25

12 Jun '25

Ben Gamari pushed new branch wip/ghc-9.14 at Glasgow Haskell Compiler / GHC -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/tree/wip/ghc-9.14 You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc] Pushed new branch wip/T26115
by Simon Peyton Jones (＠simonpj) 12 Jun '25

12 Jun '25

Simon Peyton Jones pushed new branch wip/T26115 at Glasgow Haskell Compiler / GHC -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/tree/wip/T26115 You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/T26118-remove-hptallfaminstances-usage-during-upsweep] include -hi-boot
by Patrick (＠soulomoon) 12 Jun '25

12 Jun '25

Patrick pushed to branch wip/T26118-remove-hptallfaminstances-usage-during-upsweep at Glasgow Haskell Compiler / GHC Commits: e9b8725a by soulomoon at 2025-06-12T23:56:49+08:00 include -hi-boot - - - - - 1 changed file: - compiler/GHC/Driver/Env.hs Changes: ===================================== compiler/GHC/Driver/Env.hs ===================================== @@ -232,7 +232,7 @@ hugAnnsBelow hsc uid mn = foldr (flip extendAnnEnvList) emptyAnnEnv <$> hugSomeThingsBelowUs (md_anns . hm_details) False hsc uid mn hugFamInstancesBelow :: HscEnv -> UnitId -> ModuleNameWithIsBoot -> IO (ModuleEnv FamInstEnv) -hugFamInstancesBelow = hugSomeThingsBelowUs' combine emptyModuleEnv False +hugFamInstancesBelow = hugSomeThingsBelowUs' combine emptyModuleEnv True where hmiFamInstEnv = extendFamInstEnvList emptyFamInstEnv . (md_fam_insts . hm_details) hmiModule = mi_module . hm_iface View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/e9b8725a83a80087a9d21de3944534c… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/e9b8725a83a80087a9d21de3944534c… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/T26118-remove-hptallfaminstances-usage-during-upsweep] Replace extendModuleEnv with extendModuleEnvWith for unionFamInstEnv in hugFamInstancesBelow
by Patrick (＠soulomoon) 12 Jun '25

12 Jun '25

Patrick pushed to branch wip/T26118-remove-hptallfaminstances-usage-during-upsweep at Glasgow Haskell Compiler / GHC Commits: 59a99c39 by soulomoon at 2025-06-12T23:54:56+08:00 Replace extendModuleEnv with extendModuleEnvWith for unionFamInstEnv in hugFamInstancesBelow - - - - - 1 changed file: - compiler/GHC/Driver/Env.hs Changes: ===================================== compiler/GHC/Driver/Env.hs ===================================== @@ -240,7 +240,7 @@ hugFamInstancesBelow = hugSomeThingsBelowUs' combine emptyModuleEnv False combine md acc = do let famInstEnv = hmiFamInstEnv md mod = hmiModule md - in extendModuleEnv acc mod famInstEnv + in extendModuleEnvWith unionFamInstEnv acc mod famInstEnv -- | Find all COMPLETE pragmas in modules that are in the transitive closure of the -- given module. View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/59a99c39c0ebd871e7f0122a05b7a58… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/59a99c39c0ebd871e7f0122a05b7a58… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/refactor-th-splices] Refactor the treatment of nested Template Haskell splices
by Matthew Pickering (＠mpickering) 12 Jun '25

12 Jun '25

Matthew Pickering pushed to branch wip/refactor-th-splices at Glasgow Haskell Compiler / GHC Commits: 51fc7dd4 by Matthew Pickering at 2025-06-12T16:46:33+01:00 Refactor the treatment of nested Template Haskell splices * The difference between a normal splice, a quasiquoter and implicit splice caused by lifting is stored in the AST after renaming. * Information that the renamer learns about splices is stored in the relevant splice extension points (XUntypedSpliceExpr, XQuasiQuote). * Normal splices and quasi quotes record the flavour of splice (exp/pat/dec etc) * Implicit lifting stores information about why the lift was attempted, so if it fails, that can be reported to the user. * After renaming, the decision taken to attempt to implicitly lift a variable is stored in the `XXUntypedSplice` extension field in the `HsImplicitLiftSplice` constructor. * Since all the information is stored in the AST, in `HsUntypedSplice`, the type of `PendingRnSplice` now just stores a `HsUntypedSplice`. * Error messages since the original program can be easily printed, this is noticeable in the case of implicit lifting. * The user-written syntax is directly type-checked. Before, some desugaring took place in the * Fixes .hie files to work better with nested splices (nested splices are not indexed) * The location of the quoter in a quasiquote is now located, so error messages will precisely point to it (and again, it is indexed by hie files) In the future, the typechecked AST should also retain information about the splices and the specific desugaring being left to the desugarer. Also, `runRnSplice` should call `tcUntypedSplice`, otherwise the typechecking logic is duplicated (see the `QQError` and `QQTopError` tests for a difference caused by this). - - - - - 46 changed files: - compiler/GHC/Builtin/Names/TH.hs - compiler/GHC/Hs/Expr.hs - compiler/GHC/Hs/Expr.hs-boot - compiler/GHC/Hs/Instances.hs - compiler/GHC/HsToCore/Quote.hs - compiler/GHC/Iface/Ext/Ast.hs - compiler/GHC/Parser.y - compiler/GHC/Parser/Lexer.x - compiler/GHC/Rename/Expr.hs - compiler/GHC/Rename/Splice.hs - compiler/GHC/Tc/Errors/Ppr.hs - compiler/GHC/Tc/Gen/App.hs - compiler/GHC/Tc/Gen/Expr.hs - compiler/GHC/Tc/Gen/Head.hs - compiler/GHC/Tc/Gen/Splice.hs - compiler/GHC/Tc/Gen/Splice.hs-boot - compiler/GHC/Tc/Types/ErrCtxt.hs - compiler/GHC/Tc/Types/TH.hs - compiler/GHC/Tc/Utils/Concrete.hs - compiler/GHC/Tc/Utils/TcMType.hs - compiler/GHC/ThToHs.hs - compiler/GHC/Types/Name/Reader.hs - compiler/GHC/Types/ThLevelIndex.hs - compiler/Language/Haskell/Syntax/Expr.hs - compiler/Language/Haskell/Syntax/Extension.hs - testsuite/tests/diagnostic-codes/codes.stdout - testsuite/tests/linear/should_fail/LinearTHFail.stderr - testsuite/tests/linters/notes.stdout - testsuite/tests/perf/compiler/hard_hole_fits.stderr - testsuite/tests/quasiquotation/T3953.stderr - + testsuite/tests/quotes/QQError.hs - + testsuite/tests/quotes/QQError.stderr - testsuite/tests/quotes/T10384.stderr - testsuite/tests/quotes/TH_localname.stderr - testsuite/tests/quotes/all.T - + testsuite/tests/th/QQInQuote.hs - + testsuite/tests/th/QQTopError.hs - + testsuite/tests/th/QQTopError.stderr - testsuite/tests/th/T10598_TH.stderr - testsuite/tests/th/T14681.stderr - testsuite/tests/th/T17804.stderr - testsuite/tests/th/T5508.stderr - testsuite/tests/th/TH_Lift.stderr - testsuite/tests/th/all.T - testsuite/tests/th/overloaded/TH_overloaded_constraints_fail.stderr - utils/check-exact/ExactPrint.hs The diff was not included because it is too large. View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/51fc7dd482f108042bf1714cbe606a9… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/51fc7dd482f108042bf1714cbe606a9… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/T26118-remove-hptallfaminstances-usage-during-upsweep] use unsafeInterleaveIO to only load FamInstances if needed
by Patrick (＠soulomoon) 12 Jun '25

12 Jun '25

Patrick pushed to branch wip/T26118-remove-hptallfaminstances-usage-during-upsweep at Glasgow Haskell Compiler / GHC Commits: 358a66d8 by soulomoon at 2025-06-12T23:43:39+08:00 use unsafeInterleaveIO to only load FamInstances if needed - - - - - 2 changed files: - compiler/GHC/Driver/Env.hs - compiler/GHC/Tc/Instance/Family.hs Changes: ===================================== compiler/GHC/Driver/Env.hs ===================================== @@ -266,6 +266,9 @@ hugInstancesBelow hsc_env uid mnwib = do mnwib return (foldl' unionInstEnv emptyInstEnv insts, concat famInsts) +-- | Get things from modules in the transitive closure of the given module. +-- +-- Note: Don't expose this function. This is a footgun if exposed! hugSomeThingsBelowUs' :: (HomeModInfo -> a -> a) -> a -> Bool -> HscEnv -> UnitId -> ModuleNameWithIsBoot -> IO a hugSomeThingsBelowUs' _ acc _ hsc_env _ _ | isOneShot (ghcMode (hsc_dflags hsc_env)) = return acc -- An explicit check to see if we are in one-shot mode to avoid poking the ModuleGraph thunk ===================================== compiler/GHC/Tc/Instance/Family.hs ===================================== @@ -25,6 +25,7 @@ import GHC.Core.TyCo.Rep import GHC.Core.TyCo.FVs import GHC.Iface.Load +import GHC.IO (unsafeInterleaveIO) import GHC.Tc.Errors.Types import GHC.Tc.Types.Evidence @@ -317,7 +318,7 @@ checkFamInstConsistency directlyImpMods hsc_env unitId mnwib -- See Note [Order of type family consistency checks] } - ; hpt_fam_insts <- liftIO $ hugFamInstancesBelow hsc_env unitId mnwib + ; hpt_fam_insts <- liftIO $ unsafeInterleaveIO $ hugFamInstancesBelow hsc_env unitId mnwib ; debug_consistent_set <- mapM (\x -> (\y -> (x, length y)) <$> modConsistent x) directlyImpMods ; traceTc "init_consistent_set" (ppr debug_consistent_set) ; let init_consistent_set = map fst (reverse (sortOn snd debug_consistent_set)) View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/358a66d82365f2f1b3a234adade8bc6… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/358a66d82365f2f1b3a234adade8bc6… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/refactor-th-splices] Refactor the treatment of nested Template Haskell splices
by Matthew Pickering (＠mpickering) 12 Jun '25

12 Jun '25

Matthew Pickering pushed to branch wip/refactor-th-splices at Glasgow Haskell Compiler / GHC Commits: 3e91314a by Matthew Pickering at 2025-06-12T14:37:17+01:00 Refactor the treatment of nested Template Haskell splices * The difference between a normal splice, a quasiquoter and implicit splice caused by lifting is stored in the AST after renaming. * Information that the renamer learns about splices is stored in the relevant splice extension points (XUntypedSpliceExpr, XQuasiQuote). * Normal splices and quasi quotes record the flavour of splice (exp/pat/dec etc) * Implicit lifting stores information about why the lift was attempted, so if it fails, that can be reported to the user. * After renaming, the decision taken to attempt to implicitly lift a variable is stored in the `XXUntypedSplice` extension field in the `HsImplicitLiftSplice` constructor. * Since all the information is stored in the AST, in `HsUntypedSplice`, the type of `PendingRnSplice` now just stores a `HsUntypedSplice`. * Error messages since the original program can be easily printed, this is noticeable in the case of implicit lifting. * The user-written syntax is directly type-checked. Before, some desugaring took place in the * Fixes .hie files to work better with nested splices (nested splices are not indexed) * The location of the quoter in a quasiquote is now located, so error messages will precisely point to it (and again, it is indexed by hie files) In the future, the typechecked AST should also retain information about the splices and the specific desugaring being left to the desugarer. Also, `runRnSplice` should call `tcUntypedSplice`, otherwise the typechecking logic is duplicated (see the `QQError` and `QQTopError` tests for a difference caused by this). - - - - - 46 changed files: - compiler/GHC/Builtin/Names/TH.hs - compiler/GHC/Hs/Expr.hs - compiler/GHC/Hs/Expr.hs-boot - compiler/GHC/Hs/Instances.hs - compiler/GHC/HsToCore/Quote.hs - compiler/GHC/Iface/Ext/Ast.hs - compiler/GHC/Parser.y - compiler/GHC/Parser/Lexer.x - compiler/GHC/Rename/Expr.hs - compiler/GHC/Rename/Splice.hs - compiler/GHC/Tc/Errors/Ppr.hs - compiler/GHC/Tc/Gen/App.hs - compiler/GHC/Tc/Gen/Expr.hs - compiler/GHC/Tc/Gen/Head.hs - compiler/GHC/Tc/Gen/Splice.hs - compiler/GHC/Tc/Gen/Splice.hs-boot - compiler/GHC/Tc/Types/ErrCtxt.hs - compiler/GHC/Tc/Types/TH.hs - compiler/GHC/Tc/Utils/Concrete.hs - compiler/GHC/Tc/Utils/TcMType.hs - compiler/GHC/ThToHs.hs - compiler/GHC/Types/Name/Reader.hs - compiler/GHC/Types/ThLevelIndex.hs - compiler/Language/Haskell/Syntax/Expr.hs - compiler/Language/Haskell/Syntax/Extension.hs - testsuite/tests/diagnostic-codes/codes.stdout - testsuite/tests/linear/should_fail/LinearTHFail.stderr - testsuite/tests/linters/notes.stdout - testsuite/tests/perf/compiler/hard_hole_fits.stderr - testsuite/tests/quasiquotation/T3953.stderr - + testsuite/tests/quotes/QQError.hs - + testsuite/tests/quotes/QQError.stderr - testsuite/tests/quotes/T10384.stderr - testsuite/tests/quotes/TH_localname.stderr - testsuite/tests/quotes/all.T - + testsuite/tests/th/QQInQuote.hs - + testsuite/tests/th/QQTopError.hs - + testsuite/tests/th/QQTopError.stderr - testsuite/tests/th/T10598_TH.stderr - testsuite/tests/th/T14681.stderr - testsuite/tests/th/T17804.stderr - testsuite/tests/th/T5508.stderr - testsuite/tests/th/TH_Lift.stderr - testsuite/tests/th/all.T - testsuite/tests/th/overloaded/TH_overloaded_constraints_fail.stderr - utils/check-exact/ExactPrint.hs The diff was not included because it is too large. View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/3e91314a3362f54d401be4e196a01a2… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/3e91314a3362f54d401be4e196a01a2… You're receiving this email because of your account on gitlab.haskell.org.

1 0