- ghc-commits - Haskell.org

[Git][ghc/ghc][wip/andreask/build-opt] 17 commits: Migrate `ghc-pkg` to use `OsPath` and `file-io`
by Andreas Klebinger (＠AndreasK) 20 Apr '26

20 Apr '26

Andreas Klebinger pushed to branch wip/andreask/build-opt at Glasgow Haskell Compiler / GHC Commits: 7666f4a9 by Fendor at 2026-04-17T22:29:51-04:00 Migrate `ghc-pkg` to use `OsPath` and `file-io` `ghc-pkg` should use UNC paths as much as possible to avoid MAX_PATH issues on windows. `file-io` uses UNC Paths by default on windows, ensuring we use the correct APIs and that we finally are no longer plagued by MAX_PATH issues in CI and private machines. On top of it, the higher correctness of `OsPath` is appreciated in this small codebase. Also, we improve memory usage very slightly, due to the more efficient memory representation of `OsPath` over `FilePath` Adds `ghc-pkg` regression test for MAX_PATH on windows Make sure `ghc-pkg` behaves as expected when long paths (> 255) are involved on windows. Let's generate a testcase where we can actually observe that `ghc-pkg` behaves as epxected. See the documentation for windows on Maximum Path Length Limitation: * `https://learn.microsoft.com/en-us/windows/win32/fileio/maximum-file-path-limitation` Adds changelog entry for long path support in ghc-pkg. - - - - - 78434e8c by Simon Peyton Jones at 2026-04-17T22:30:38-04:00 Kill off the substitution in Lint Now that we have invariant (NoTypeShadowing) we no longer need Lint to carry an ambient substitution. This makes it simpler and faster. A really worthwhile refactor. There are some knock-on effects * Linting join points after worker/wrapper. See Note [Join points and beta redexes] * Running a type substitution after the desugarer. See Note [Substituting type-lets] in the new module GHC.Core.SubstTypeLets Implements #27078 Most perf tests don't use Lint so we won't see a perf incresae. But T1969, which uses -O0 and Lint, gets 1.3% worse because it has to run the SubstTypeLets pass which is a somewhat expensive no-op Overall though compile-time allocations are down 0.1%. Metric Increase: T1969 - - - - - 86ca6c2c by mangoiv at 2026-04-17T22:31:22-04:00 testsuite: inline elemCoreTest Some weird (probably python scoping) rule caused elemCoreTest, a regex being out of scope on ubuntu, presumably because of a newer python version. This patch just inlines the regex, which fixes the issue. Fixes #27193 - - - - - 1f30005b by Andreas Klebinger at 2026-04-19T21:48:05+00:00 SOURCE import parseIdentifer to improve parallelism. - - - - - e96b8d63 by Andreas Klebinger at 2026-04-19T21:48:05+00:00 CallerCC: Remove CoreM dependency Pass DynFlags directly, making the dependency graph more parallel. - - - - - 26090197 by Andreas Klebinger at 2026-04-19T21:58:28+00:00 SOURCE-import HsInstances inside ghc. Fixes #27198 by SOURCE importing HsInstances unlocking more build paralleism. (cherry picked from commit 7e53835977538e3c37e5d78502ff422ad87d4f9d) - - - - - cb679710 by Andreas Klebinger at 2026-04-19T22:06:08+00:00 HsInstances - fix a warning - - - - - f5722894 by Andreas Klebinger at 2026-04-20T15:11:06+00:00 SOURCE import GHC.Types.Error in some places for build parallelism. Performance for these interfaces is not very relevant since it's only used for error handling. This means we can use SOURCE imports to shorten the critical build path by a non trivial amount. - - - - - a64cd962 by Andreas Klebinger at 2026-04-20T15:11:26+00:00 Split GHC.Driver.Main.hs up into multiple components. This module was getting far too large to reason about, it split it into components that all are re-exported from GHC.Driver.Main - - - - - cf62c053 by Andreas Klebinger at 2026-04-20T15:42:58+00:00 Fix hs-boot import - - - - - e6550ad1 by Andreas Klebinger at 2026-04-20T15:45:33+00:00 Fix hs-boot import2 - - - - - 900317b2 by Andreas Klebinger at 2026-04-20T15:54:23+00:00 Use SOURCE header for Driver.Main.Compiler import - - - - - 8de7314b by Andreas Klebinger at 2026-04-20T16:23:24+00:00 more hs-boot workarounds - - - - - c6ce86ca by Andreas Klebinger at 2026-04-20T17:15:07+00:00 Driver interactive boot import - - - - - 19fd9f7e by Andreas Klebinger at 2026-04-20T17:18:27+00:00 deSugar SOURCE header - - - - - eff8fd32 by Andreas Klebinger at 2026-04-20T17:27:00+00:00 Add GHC.Tc.Deriv.hs-boot - - - - - ee3550f9 by Andreas Klebinger at 2026-04-20T18:56:35+00:00 Remove a potentially pointless import - - - - - 44 changed files: - + changelog.d/ghc-pkg-long-path-support - compiler/GHC/Core/Lint.hs - + compiler/GHC/Core/Lint/SubstTypeLets.hs - compiler/GHC/Core/Opt/CallerCC.hs - compiler/GHC/Core/Opt/OccurAnal.hs - compiler/GHC/Core/Opt/Pipeline.hs - compiler/GHC/Core/Opt/Stats.hs - compiler/GHC/Core/Opt/WorkWrap/Utils.hs - compiler/GHC/Core/Subst.hs - compiler/GHC/Driver/Backpack.hs - compiler/GHC/Driver/Config/Core/Lint.hs - compiler/GHC/Driver/Env/Types.hs - compiler/GHC/Driver/Errors/Types.hs - compiler/GHC/Driver/Main.hs - + compiler/GHC/Driver/Main/Compile.hs - compiler/GHC/Driver/Main.hs-boot → compiler/GHC/Driver/Main/Compile.hs-boot - + compiler/GHC/Driver/Main/Hsc.hs - + compiler/GHC/Driver/Main/Interactive.hs - + compiler/GHC/Driver/Main/Passes.hs - + compiler/GHC/Driver/Main/Passes.hs-boot - compiler/GHC/Driver/Session.hs - compiler/GHC/Hs.hs - compiler/GHC/Hs/Instances.hs - + compiler/GHC/Hs/Instances.hs-boot - + compiler/GHC/HsToCore.hs-boot - compiler/GHC/Iface/Load.hs - + compiler/GHC/Tc/Deriv.hs-boot - compiler/GHC/Tc/TyCl.hs - compiler/GHC/Tc/TyCl/Instance.hs - compiler/GHC/Tc/TyCl/Instance.hs-boot - compiler/GHC/Types/Error.hs - compiler/GHC/Types/Error.hs-boot - compiler/GHC/Unit/State.hs - compiler/ghc.cabal.in - libraries/base/tests/perf/all.T - libraries/ghc-boot/GHC/Unit/Database.hs - libraries/ghc-boot/ghc-boot.cabal.in - testsuite/tests/cabal/Makefile - testsuite/tests/cabal/all.T - + testsuite/tests/cabal/ghcpkg10.stdout - testsuite/tests/corelint/LintEtaExpand.stderr - testsuite/tests/corelint/T21115b.stderr - utils/ghc-pkg/Main.hs - utils/ghc-pkg/ghc-pkg.cabal.in The diff was not included because it is too large. View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/3c622f0df932d3d0e0a36ec171499a… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/3c622f0df932d3d0e0a36ec171499a… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/spj-reinstallable-base2] Wibbles
by Simon Peyton Jones (＠simonpj) 20 Apr '26

20 Apr '26

Simon Peyton Jones pushed to branch wip/spj-reinstallable-base2 at Glasgow Haskell Compiler / GHC Commits: 9af7210e by Simon Peyton Jones at 2026-04-20T17:41:41+01:00 Wibbles ... around (>>>) and Floating class - - - - - 6 changed files: - compiler/GHC/Builtin/KnownOccs.hs - compiler/GHC/Builtin/TH.hs - compiler/GHC/HsToCore/Monad.hs - compiler/GHC/HsToCore/Quote.hs - compiler/GHC/Iface/Errors/Ppr.hs - libraries/base/src/GHC/KnownKeyNames.hs Changes: ===================================== compiler/GHC/Builtin/KnownOccs.hs ===================================== @@ -86,6 +86,11 @@ have the correct one in scope when looking up a known-occ name. module GHC.Internal.Base where foldrList = foldr make `foldrList` known-occ, and refer to that in desugaring list comprehensions. + +* (>>>). You might think that the known-occ version is the one defined in + GHC.Internal.Control.Category. But no, it isn't. We have a different one + (albeit with the same definitino!) in GHC.Internal.Desugar, whose type + has the right "shape" type for `newKnownOccMethod`. Sigh. -} ===================================== compiler/GHC/Builtin/TH.hs ===================================== @@ -903,11 +903,6 @@ forallEIdKey = mkPreludeMiscIdUnique 802 forallVisEIdKey = mkPreludeMiscIdUnique 803 constrainedEIdKey = mkPreludeMiscIdUnique 804 --- data Guard = ... -normalGEIdKey, patGEIdKey :: Unique -normalGEIdKey = mkPreludeMiscIdUnique 310 -patGEIdKey = mkPreludeMiscIdUnique 311 - -- data Dec = ... funDIdKey, valDIdKey, dataDIdKey, newtypeDIdKey, tySynDIdKey, classDIdKey, instanceWithOverlapDIdKey, instanceDIdKey, sigDIdKey, forImpDIdKey, ===================================== compiler/GHC/HsToCore/Monad.hs ===================================== @@ -629,7 +629,7 @@ dsLookupKnownKeyName uniq Succeeded name -> return name Failed msg -> failIfM (pprDiagnostic msg) } } -dsLookupKnownKeyThing :: KnownKey -> DsM TyThing +dsLookupKnownKeyThing :: HasDebugCallStack => KnownKey -> DsM TyThing dsLookupKnownKeyThing uniq = do { rebindable_src <- dsGetKnownKeySource ; dsToIfL $ @@ -638,13 +638,13 @@ dsLookupKnownKeyThing uniq Succeeded thing -> return thing Failed msg -> failIfM (pprDiagnostic msg) } } -dsLookupKnownKeyTyCon :: KnownKey -> DsM TyCon +dsLookupKnownKeyTyCon :: HasDebugCallStack => KnownKey -> DsM TyCon dsLookupKnownKeyTyCon uniq = tyThingTyCon <$> dsLookupKnownKeyThing uniq -dsLookupKnownKeyDataCon :: KnownKey -> DsM DataCon +dsLookupKnownKeyDataCon :: HasDebugCallStack => KnownKey -> DsM DataCon dsLookupKnownKeyDataCon uniq = tyThingDataCon <$> dsLookupKnownKeyThing uniq -dsLookupKnownKeyId :: KnownKey -> DsM Id +dsLookupKnownKeyId :: HasDebugCallStack => KnownKey -> DsM Id dsLookupKnownKeyId uniq = tyThingId <$> dsLookupKnownKeyThing uniq -------------------------------------- ===================================== compiler/GHC/HsToCore/Quote.hs ===================================== @@ -2370,10 +2370,10 @@ lookupType :: KnownOcc -- Name of type constructor (e.g. (M TH.Exp)) lookupType tc_name = do { tc <- lift $ dsLookupKnownOccTyCon tc_name ; return (mkTyConTy tc) } -lookupKnownKeyType :: Unique -- Unique of type constructor (e.g. (M TH.Exp)) +lookupKnownOccType :: KnownOcc -- Occ-name of type constructor (e.g. (M TH.Exp)) -> MetaM Type -- The type -lookupKnownKeyType tc_key - = do { tc <- lift $ dsLookupKnownKeyTyCon tc_key +lookupKnownOccType tc_key + = do { tc <- lift $ dsLookupKnownOccTyCon tc_key ; return (mkTyConApp tc []) } wrapGenSyms :: [GenSymBind] @@ -3133,7 +3133,7 @@ mk_integer :: Integer -> MetaM (HsLit GhcTc) mk_integer i = return $ XLit $ HsInteger NoSourceText i integerTy mk_rational :: FractionalLit -> MetaM (HsLit GhcTc) -mk_rational r = do rat_ty <- lookupKnownKeyType rationalTyConKey +mk_rational r = do rat_ty <- lookupKnownOccType rationalTyConOcc return $ XLit $ HsRat r rat_ty mk_string :: FastString -> MetaM (HsLit GhcRn) ===================================== compiler/GHC/Iface/Errors/Ppr.hs ===================================== @@ -299,7 +299,8 @@ interfaceErrorDiagnostic opts = \ case MissingKnownKey1 key -> hang (text "Could not find known key" <+> quotes (pprKnownKey key)) 2 (vcat [ text "in the exports of GHC.KnownKeys" - , text "occname:" <+> pp_occ (knownKeyOccName_maybe key) ]) + , text "occname:" <+> pp_occ (knownKeyOccName_maybe key) + , text "REMEMBER: for tycons, divide by 2!!"]) where pp_occ (Just occ) = ppr occ pp_occ Nothing = text "Yikes: that key isn't in the known-key table" ===================================== libraries/base/src/GHC/KnownKeyNames.hs ===================================== @@ -92,7 +92,7 @@ module GHC.KnownKeyNames , error -- Numbers - , Num, Integral, Real, Fractional, RealFloat, RealFrac + , Num, Integral, Real, Floating, Fractional, RealFloat, RealFrac , (+), (-), (*), negate, fromInteger , divInt#, modInt# @@ -204,6 +204,7 @@ module GHC.KnownKeyNames , Clause, clause , Stmt, bindS, letS, noBindS, parS, recS , Body, normalB, guardedB + , Guard, normalGE, patGE ) where import GHC.Internal.Base hiding( foldr ) @@ -226,7 +227,8 @@ import GHC.Internal.IO( seq# ) import GHC.Internal.Control.Monad( fail, guard ) import GHC.Internal.Control.Monad.Fix( mfix, loop ) import GHC.Internal.Control.Monad.Zip( mzip ) -import GHC.Internal.Control.Arrow( arr, (>>>), first, app, (|||) ) +import GHC.Internal.Control.Arrow( arr, first, app, (|||) ) +import GHC.Internal.Desugar( (>>>) ) -- See Note [Tricky known-occ cases] import GHC.Internal.OverloadedLabels( fromLabel ) import GHC.Internal.Records import GHC.Internal.CString as CS View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/9af7210e9c2adf040bf4ea4466c6035… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/9af7210e9c2adf040bf4ea4466c6035… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/marge_bot_batch_merge_job] 3 commits: NCG: Implement constant folding for vector simd ops (Issue #25030)
by Marge Bot (＠marge-bot) 20 Apr '26

20 Apr '26

Marge Bot pushed to branch wip/marge_bot_batch_merge_job at Glasgow Haskell Compiler / GHC Commits: 311c6d60 by aparker at 2026-04-20T11:33:14-04:00 NCG: Implement constant folding for vector simd ops (Issue #25030) - - - - - e5743d44 by sheaf at 2026-04-20T11:33:14-04:00 Mark some SIMD tests as broken on i386 optllvm As seen in #25498, several SIMD tests are broken on i386 in the optllvm way. This commit marks them as "expect_broken". - - - - - 82d2f00e by Wolfgang Jeltsch at 2026-04-20T11:33:15-04:00 Move most of the `System.IO` implementation into `base` This involves a rewrite of the `combine` helper function to avoid the use of `last`, which would now be flagged as an error. Metric Decrease: LinkableUsage01 T3294 Metric Increase: T12227 T12707 T5642 - - - - - 34 changed files: - + changelog.d/simd_constant_folding - compiler/GHC/Cmm/Opt.hs - compiler/GHC/Utils/Misc.hs - libraries/base/src/Control/Concurrent.hs - libraries/base/src/GHC/IO/Handle.hs - libraries/base/src/Prelude.hs - libraries/base/src/System/IO.hs - libraries/base/src/Text/Printf.hs - libraries/ghc-internal/src/GHC/Internal/System/IO.hs - 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/simd/should_run/Makefile - + testsuite/tests/simd/should_run/T25030.hs - + testsuite/tests/simd/should_run/T25030.stdout - testsuite/tests/simd/should_run/all.T - testsuite/tests/typecheck/should_compile/T9497a.stderr - testsuite/tests/typecheck/should_compile/holes.stderr - testsuite/tests/typecheck/should_compile/holes3.stderr - testsuite/tests/typecheck/should_compile/valid_hole_fits.stderr - testsuite/tests/typecheck/should_fail/T9497d.stderr - testsuite/tests/typecheck/should_run/T9497a-run.stderr - testsuite/tests/typecheck/should_run/T9497b-run.stderr - testsuite/tests/typecheck/should_run/T9497c-run.stderr - utils/haddock/html-test/ref/Bug1004.html - utils/haddock/html-test/ref/Bug973.html - utils/haddock/html-test/ref/ConstructorPatternExport.html - utils/haddock/html-test/ref/DefaultSignatures.html - utils/haddock/html-test/ref/Hash.html - utils/haddock/html-test/ref/PatternSyns.html - utils/haddock/html-test/ref/PatternSyns2.html - utils/haddock/html-test/ref/QuasiExpr.html - utils/haddock/html-test/ref/Test.html The diff was not included because it is too large. View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/6d2aa4ad60064eefc2156b369120c2… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/6d2aa4ad60064eefc2156b369120c2… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/marge_bot_batch_merge_job] 3 commits: NCG: Implement constant folding for vector simd ops (Issue #25030)
by Marge Bot (＠marge-bot) 20 Apr '26

20 Apr '26

Marge Bot pushed to branch wip/marge_bot_batch_merge_job at Glasgow Haskell Compiler / GHC Commits: c236fd68 by aparker at 2026-04-20T07:27:47-04:00 NCG: Implement constant folding for vector simd ops (Issue #25030) - - - - - 6665c649 by sheaf at 2026-04-20T07:27:48-04:00 Mark some SIMD tests as broken on i386 optllvm As seen in #25498, several SIMD tests are broken on i386 in the optllvm way. This commit marks them as "expect_broken". - - - - - 6d2aa4ad by Wolfgang Jeltsch at 2026-04-20T07:27:49-04:00 Move most of the `System.IO` implementation into `base` This involves a rewrite of the `combine` helper function to avoid the use of `last`, which would now be flagged as an error. Metric Decrease: LinkableUsage01 T3294 Metric Increase: T12227 T12707 T5642 - - - - - 34 changed files: - + changelog.d/simd_constant_folding - compiler/GHC/Cmm/Opt.hs - compiler/GHC/Utils/Misc.hs - libraries/base/src/Control/Concurrent.hs - libraries/base/src/GHC/IO/Handle.hs - libraries/base/src/Prelude.hs - libraries/base/src/System/IO.hs - libraries/base/src/Text/Printf.hs - libraries/ghc-internal/src/GHC/Internal/System/IO.hs - 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/simd/should_run/Makefile - + testsuite/tests/simd/should_run/T25030.hs - + testsuite/tests/simd/should_run/T25030.stdout - testsuite/tests/simd/should_run/all.T - testsuite/tests/typecheck/should_compile/T9497a.stderr - testsuite/tests/typecheck/should_compile/holes.stderr - testsuite/tests/typecheck/should_compile/holes3.stderr - testsuite/tests/typecheck/should_compile/valid_hole_fits.stderr - testsuite/tests/typecheck/should_fail/T9497d.stderr - testsuite/tests/typecheck/should_run/T9497a-run.stderr - testsuite/tests/typecheck/should_run/T9497b-run.stderr - testsuite/tests/typecheck/should_run/T9497c-run.stderr - utils/haddock/html-test/ref/Bug1004.html - utils/haddock/html-test/ref/Bug973.html - utils/haddock/html-test/ref/ConstructorPatternExport.html - utils/haddock/html-test/ref/DefaultSignatures.html - utils/haddock/html-test/ref/Hash.html - utils/haddock/html-test/ref/PatternSyns.html - utils/haddock/html-test/ref/PatternSyns2.html - utils/haddock/html-test/ref/QuasiExpr.html - utils/haddock/html-test/ref/Test.html The diff was not included because it is too large. View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/bbf4abb9729bfa017faacd4d9ea673… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/bbf4abb9729bfa017faacd4d9ea673… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/marge_bot_batch_merge_job] 3 commits: NCG: Implement constant folding for vector simd ops (Issue #25030)
by Marge Bot (＠marge-bot) 20 Apr '26

20 Apr '26

Marge Bot pushed to branch wip/marge_bot_batch_merge_job at Glasgow Haskell Compiler / GHC Commits: 42fba2df by aparker at 2026-04-20T00:57:05-04:00 NCG: Implement constant folding for vector simd ops (Issue #25030) - - - - - 0962dc2a by sheaf at 2026-04-20T00:57:06-04:00 Mark some SIMD tests as broken on i386 optllvm As seen in #25498, several SIMD tests are broken on i386 in the optllvm way. This commit marks them as "expect_broken". - - - - - bbf4abb9 by Wolfgang Jeltsch at 2026-04-20T00:57:07-04:00 Move most of the `System.IO` implementation into `base` This involves a rewrite of the `combine` helper function to avoid the use of `last`, which would now be flagged as an error. Metric Decrease: LinkableUsage01 T3294 Metric Increase: T12227 T12707 T5642 - - - - - 34 changed files: - + changelog.d/simd_constant_folding - compiler/GHC/Cmm/Opt.hs - compiler/GHC/Utils/Misc.hs - libraries/base/src/Control/Concurrent.hs - libraries/base/src/GHC/IO/Handle.hs - libraries/base/src/Prelude.hs - libraries/base/src/System/IO.hs - libraries/base/src/Text/Printf.hs - libraries/ghc-internal/src/GHC/Internal/System/IO.hs - 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/simd/should_run/Makefile - + testsuite/tests/simd/should_run/T25030.hs - + testsuite/tests/simd/should_run/T25030.stdout - testsuite/tests/simd/should_run/all.T - testsuite/tests/typecheck/should_compile/T9497a.stderr - testsuite/tests/typecheck/should_compile/holes.stderr - testsuite/tests/typecheck/should_compile/holes3.stderr - testsuite/tests/typecheck/should_compile/valid_hole_fits.stderr - testsuite/tests/typecheck/should_fail/T9497d.stderr - testsuite/tests/typecheck/should_run/T9497a-run.stderr - testsuite/tests/typecheck/should_run/T9497b-run.stderr - testsuite/tests/typecheck/should_run/T9497c-run.stderr - utils/haddock/html-test/ref/Bug1004.html - utils/haddock/html-test/ref/Bug973.html - utils/haddock/html-test/ref/ConstructorPatternExport.html - utils/haddock/html-test/ref/DefaultSignatures.html - utils/haddock/html-test/ref/Hash.html - utils/haddock/html-test/ref/PatternSyns.html - utils/haddock/html-test/ref/PatternSyns2.html - utils/haddock/html-test/ref/QuasiExpr.html - utils/haddock/html-test/ref/Test.html The diff was not included because it is too large. View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/371ac8b6d4904a11214229b4c0d4d1… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/371ac8b6d4904a11214229b4c0d4d1… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/ani/T27156] HsQualLit via Expand
by Apoorv Ingle (＠ani) 20 Apr '26

20 Apr '26

Apoorv Ingle pushed to branch wip/ani/T27156 at Glasgow Haskell Compiler / GHC Commits: c6ae25eb by Apoorv Ingle at 2026-04-19T20:59:43-05:00 HsQualLit via Expand - - - - - 3 changed files: - compiler/GHC/Rename/Expr.hs - compiler/GHC/Rename/Utils.hs - compiler/GHC/Tc/Gen/Expand.hs Changes: ===================================== compiler/GHC/Rename/Expr.hs ===================================== @@ -40,7 +40,6 @@ import GHC.Rename.Unbound ( reportUnboundName ) import GHC.Rename.Splice ( rnTypedBracket, rnUntypedBracket, rnTypedSplice , rnUntypedSpliceExpr, checkThLocalNameWithLift, checkThLocalNameNoLift ) import GHC.Rename.HsType -import GHC.Rename.Lit import GHC.Rename.Pat import GHC.Driver.DynFlags @@ -369,10 +368,11 @@ rnExpr (HsOverLit x lit) return (HsApp noExtField (noLocA neg) (noLocA (HsOverLit x lit')) , fvs ) } -rnExpr (HsQualLit x lit) = do - ((lit', desugaredExpr), fvs) <- rnQualLit lit - let origExpr = HsQualLit x lit' - return (mkExpandedExpr origExpr desugaredExpr, fvs) +rnExpr (HsQualLit x QualLit{..}) + = do { (funName, fvs) <- lookupNameWithQualifier fromStringName ql_mod + ; let lit' = QualLit{ql_ext = L noAnn funName, ..} + ; return (HsQualLit x lit', fvs) + } rnExpr (HsApp x fun arg) = do { (fun',fvFun) <- rnLExpr fun ===================================== compiler/GHC/Rename/Utils.hs ===================================== @@ -41,7 +41,7 @@ module GHC.Rename.Utils ( where -import GHC.Prelude hiding (init, last, scanl, tail) +import GHC.Prelude import GHC.Core.Type import GHC.Hs ===================================== compiler/GHC/Tc/Gen/Expand.hs ===================================== @@ -131,6 +131,20 @@ tcExpand e@(HsOverLabel (_, rs_table) v) | otherwise = pprPanic "tcExpand" (vcat [ text "Should Never Happen: could not find fromLabel in rs_table" , ppr e] ) + +------------------------------------------ +-- Qualified Literals +tcExpand e@(HsQualLit _ QualLit{ql_val = ql_val, ql_ext = (L _ fromStringName)}) + = do { let hsLit = case ql_val of + -- See Note [Implementation of QualifiedStrings] + HsQualString st s -> HsString st s + ; return $ Just $ + HSE { hse_ctxt = ExprCtxt e + , hse_exp = wrapGenSpan $ genHsApps fromStringName [genLHsLit hsLit] + } + } + + ------------------------------------------ -- Operator Applications tcExpand e@(OpApp _ arg1 op arg2) @@ -186,7 +200,7 @@ tcExpand (HsDo _ do_or_lc stmts) -- ApplicativeDo are typechecked using tcDoStmts = do isApplicativeDo <- xoptM LangExt.ApplicativeDo if isApplicativeDo - then return Nothing + then return Nothing -- to be fixed by #24406 -- Expand expression on the fly otherwise -- See Note [Typechecking by expansion: overview] else do { hse <- expandDoStmts do_or_lc stmts @@ -254,10 +268,9 @@ tcExpand e@(RecordUpd (Just rs_table) (L l expr) (OverloadedRecUpdFields { olRec ---------- --- We return ExpandedThingRn as is for now, --- but after removing all the calls to mkExpandExpr in the renamer, --- this case should never happen, as the renamer will never produce an ExpandedThingRn +------------------------ +-- XExpr +-- Expansions are idempotent, XExprs do not expand again tcExpand (XExpr (ExpandedThingRn hse)) = return (Just hse) View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/c6ae25ebd38670897679bdaf24f1c59… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/c6ae25ebd38670897679bdaf24f1c59… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/ani/T27156] HsQualLit via Expand
by Apoorv Ingle (＠ani) 20 Apr '26

20 Apr '26

Apoorv Ingle pushed to branch wip/ani/T27156 at Glasgow Haskell Compiler / GHC Commits: 20ed7c20 by Apoorv Ingle at 2026-04-19T20:24:26-05:00 HsQualLit via Expand - - - - - 2 changed files: - compiler/GHC/Rename/Expr.hs - compiler/GHC/Tc/Gen/Expand.hs Changes: ===================================== compiler/GHC/Rename/Expr.hs ===================================== @@ -40,7 +40,6 @@ import GHC.Rename.Unbound ( reportUnboundName ) import GHC.Rename.Splice ( rnTypedBracket, rnUntypedBracket, rnTypedSplice , rnUntypedSpliceExpr, checkThLocalNameWithLift, checkThLocalNameNoLift ) import GHC.Rename.HsType -import GHC.Rename.Lit import GHC.Rename.Pat import GHC.Driver.DynFlags @@ -369,10 +368,18 @@ rnExpr (HsOverLit x lit) return (HsApp noExtField (noLocA neg) (noLocA (HsOverLit x lit')) , fvs ) } -rnExpr (HsQualLit x lit) = do - ((lit', desugaredExpr), fvs) <- rnQualLit lit - let origExpr = HsQualLit x lit' - return (mkExpandedExpr origExpr desugaredExpr, fvs) +rnExpr (HsQualLit x lit) + = do { (lit', fvs) <- rnQualLit lit + ; return (HsQualLit x lit', fvs) + } + + where + rnQualLit :: HsQualLit GhcPs -> RnM (HsQualLit GhcRn, FreeNames) + rnQualLit QualLit{..} + = do { (funName, fvs) <- lookupNameWithQualifier fromStringName ql_mod + ; let lit = QualLit{ql_ext = L noAnn funName, ..} + ; pure (lit, fvs) } + rnExpr (HsApp x fun arg) = do { (fun',fvFun) <- rnLExpr fun ===================================== compiler/GHC/Tc/Gen/Expand.hs ===================================== @@ -131,6 +131,20 @@ tcExpand e@(HsOverLabel (_, rs_table) v) | otherwise = pprPanic "tcExpand" (vcat [ text "Should Never Happen: could not find fromLabel in rs_table" , ppr e] ) + +------------------------------------------ +-- Qualified Literals +tcExpand e@(HsQualLit _ QualLit{ql_val = ql_val, ql_ext = (L _ fromStringName)}) + = do { let hsLit = case ql_val of + -- See Note [Implementation of QualifiedStrings] + HsQualString st s -> HsString st s + ; return $ Just $ + HSE { hse_ctxt = ExprCtxt e + , hse_exp = wrapGenSpan $ genHsApps fromStringName [genLHsLit hsLit] + } + } + + ------------------------------------------ -- Operator Applications tcExpand e@(OpApp _ arg1 op arg2) View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/20ed7c203b886f44541defd1eefee74… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/20ed7c203b886f44541defd1eefee74… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/marge_bot_batch_merge_job] 6 commits: Migrate `ghc-pkg` to use `OsPath` and `file-io`
by Marge Bot (＠marge-bot) 20 Apr '26

20 Apr '26

Marge Bot pushed to branch wip/marge_bot_batch_merge_job at Glasgow Haskell Compiler / GHC Commits: 7666f4a9 by Fendor at 2026-04-17T22:29:51-04:00 Migrate `ghc-pkg` to use `OsPath` and `file-io` `ghc-pkg` should use UNC paths as much as possible to avoid MAX_PATH issues on windows. `file-io` uses UNC Paths by default on windows, ensuring we use the correct APIs and that we finally are no longer plagued by MAX_PATH issues in CI and private machines. On top of it, the higher correctness of `OsPath` is appreciated in this small codebase. Also, we improve memory usage very slightly, due to the more efficient memory representation of `OsPath` over `FilePath` Adds `ghc-pkg` regression test for MAX_PATH on windows Make sure `ghc-pkg` behaves as expected when long paths (> 255) are involved on windows. Let's generate a testcase where we can actually observe that `ghc-pkg` behaves as epxected. See the documentation for windows on Maximum Path Length Limitation: * `https://learn.microsoft.com/en-us/windows/win32/fileio/maximum-file-path-limitation` Adds changelog entry for long path support in ghc-pkg. - - - - - 78434e8c by Simon Peyton Jones at 2026-04-17T22:30:38-04:00 Kill off the substitution in Lint Now that we have invariant (NoTypeShadowing) we no longer need Lint to carry an ambient substitution. This makes it simpler and faster. A really worthwhile refactor. There are some knock-on effects * Linting join points after worker/wrapper. See Note [Join points and beta redexes] * Running a type substitution after the desugarer. See Note [Substituting type-lets] in the new module GHC.Core.SubstTypeLets Implements #27078 Most perf tests don't use Lint so we won't see a perf incresae. But T1969, which uses -O0 and Lint, gets 1.3% worse because it has to run the SubstTypeLets pass which is a somewhat expensive no-op Overall though compile-time allocations are down 0.1%. Metric Increase: T1969 - - - - - 86ca6c2c by mangoiv at 2026-04-17T22:31:22-04:00 testsuite: inline elemCoreTest Some weird (probably python scoping) rule caused elemCoreTest, a regex being out of scope on ubuntu, presumably because of a newer python version. This patch just inlines the regex, which fixes the issue. Fixes #27193 - - - - - ad87d775 by aparker at 2026-04-19T20:44:18-04:00 NCG: Implement constant folding for vector simd ops (Issue #25030) - - - - - 1480d0e6 by sheaf at 2026-04-19T20:44:19-04:00 Mark some SIMD tests as broken on i386 optllvm As seen in #25498, several SIMD tests are broken on i386 in the optllvm way. This commit marks them as "expect_broken". - - - - - 371ac8b6 by Wolfgang Jeltsch at 2026-04-19T20:44:20-04:00 Move most of the `System.IO` implementation into `base` This involves a rewrite of the `combine` helper function to avoid the use of `last`, which would now be flagged as an error. Metric Decrease: LinkableUsage01 T3294 Metric Increase: T12227 T12707 T5642 - - - - - 53 changed files: - + changelog.d/ghc-pkg-long-path-support - + changelog.d/simd_constant_folding - compiler/GHC/Cmm/Opt.hs - compiler/GHC/Core/Lint.hs - + compiler/GHC/Core/Lint/SubstTypeLets.hs - compiler/GHC/Core/Opt/OccurAnal.hs - compiler/GHC/Core/Opt/WorkWrap/Utils.hs - compiler/GHC/Core/Subst.hs - compiler/GHC/Driver/Config/Core/Lint.hs - compiler/GHC/Unit/State.hs - compiler/GHC/Utils/Misc.hs - compiler/ghc.cabal.in - libraries/base/src/Control/Concurrent.hs - libraries/base/src/GHC/IO/Handle.hs - libraries/base/src/Prelude.hs - libraries/base/src/System/IO.hs - libraries/base/src/Text/Printf.hs - libraries/base/tests/perf/all.T - libraries/ghc-boot/GHC/Unit/Database.hs - libraries/ghc-boot/ghc-boot.cabal.in - libraries/ghc-internal/src/GHC/Internal/System/IO.hs - testsuite/tests/cabal/Makefile - testsuite/tests/cabal/all.T - + testsuite/tests/cabal/ghcpkg10.stdout - testsuite/tests/corelint/LintEtaExpand.stderr - testsuite/tests/corelint/T21115b.stderr - 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/simd/should_run/Makefile - + testsuite/tests/simd/should_run/T25030.hs - + testsuite/tests/simd/should_run/T25030.stdout - testsuite/tests/simd/should_run/all.T - testsuite/tests/typecheck/should_compile/T9497a.stderr - testsuite/tests/typecheck/should_compile/holes.stderr - testsuite/tests/typecheck/should_compile/holes3.stderr - testsuite/tests/typecheck/should_compile/valid_hole_fits.stderr - testsuite/tests/typecheck/should_fail/T9497d.stderr - testsuite/tests/typecheck/should_run/T9497a-run.stderr - testsuite/tests/typecheck/should_run/T9497b-run.stderr - testsuite/tests/typecheck/should_run/T9497c-run.stderr - utils/ghc-pkg/Main.hs - utils/ghc-pkg/ghc-pkg.cabal.in - utils/haddock/html-test/ref/Bug1004.html - utils/haddock/html-test/ref/Bug973.html - utils/haddock/html-test/ref/ConstructorPatternExport.html - utils/haddock/html-test/ref/DefaultSignatures.html - utils/haddock/html-test/ref/Hash.html - utils/haddock/html-test/ref/PatternSyns.html - utils/haddock/html-test/ref/PatternSyns2.html - utils/haddock/html-test/ref/QuasiExpr.html - utils/haddock/html-test/ref/Test.html The diff was not included because it is too large. View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/36a6604123b7920f6e67fe285cf018… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/36a6604123b7920f6e67fe285cf018… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/ani/T27156] 7 commits: Migrate `ghc-pkg` to use `OsPath` and `file-io`
by Apoorv Ingle (＠ani) 20 Apr '26

20 Apr '26

Apoorv Ingle pushed to branch wip/ani/T27156 at Glasgow Haskell Compiler / GHC Commits: 7666f4a9 by Fendor at 2026-04-17T22:29:51-04:00 Migrate `ghc-pkg` to use `OsPath` and `file-io` `ghc-pkg` should use UNC paths as much as possible to avoid MAX_PATH issues on windows. `file-io` uses UNC Paths by default on windows, ensuring we use the correct APIs and that we finally are no longer plagued by MAX_PATH issues in CI and private machines. On top of it, the higher correctness of `OsPath` is appreciated in this small codebase. Also, we improve memory usage very slightly, due to the more efficient memory representation of `OsPath` over `FilePath` Adds `ghc-pkg` regression test for MAX_PATH on windows Make sure `ghc-pkg` behaves as expected when long paths (> 255) are involved on windows. Let's generate a testcase where we can actually observe that `ghc-pkg` behaves as epxected. See the documentation for windows on Maximum Path Length Limitation: * `https://learn.microsoft.com/en-us/windows/win32/fileio/maximum-file-path-limitation` Adds changelog entry for long path support in ghc-pkg. - - - - - 78434e8c by Simon Peyton Jones at 2026-04-17T22:30:38-04:00 Kill off the substitution in Lint Now that we have invariant (NoTypeShadowing) we no longer need Lint to carry an ambient substitution. This makes it simpler and faster. A really worthwhile refactor. There are some knock-on effects * Linting join points after worker/wrapper. See Note [Join points and beta redexes] * Running a type substitution after the desugarer. See Note [Substituting type-lets] in the new module GHC.Core.SubstTypeLets Implements #27078 Most perf tests don't use Lint so we won't see a perf incresae. But T1969, which uses -O0 and Lint, gets 1.3% worse because it has to run the SubstTypeLets pass which is a somewhat expensive no-op Overall though compile-time allocations are down 0.1%. Metric Increase: T1969 - - - - - 86ca6c2c by mangoiv at 2026-04-17T22:31:22-04:00 testsuite: inline elemCoreTest Some weird (probably python scoping) rule caused elemCoreTest, a regex being out of scope on ubuntu, presumably because of a newer python version. This patch just inlines the regex, which fixes the issue. Fixes #27193 - - - - - 1f810bb3 by Apoorv Ingle at 2026-04-19T19:35:25-05:00 move SectionL and SectionR into tcExpand Work on #27156 - - - - - e211351c by Apoorv Ingle at 2026-04-19T19:35:25-05:00 move Record dot syntax expansion from renamer to Expand - - - - - 591d494d by Apoorv Ingle at 2026-04-19T19:35:25-05:00 tcExpr to do expansion first, then type checking - - - - - c93a3b91 by Apoorv Ingle at 2026-04-19T19:35:25-05:00 - Add RebindableSyntaxTable to store function/operator names while renaming in the XBlah fields - Expand the following expressins right before typechecking instead of in the renamer * OverloadedLabel * RecordDotSyntax: HsGetField and HsProjection * ExplicitList * RecordUpd * HsIf - - - - - 28 changed files: - + changelog.d/ghc-pkg-long-path-support - compiler/GHC/Core/Lint.hs - + compiler/GHC/Core/Lint/SubstTypeLets.hs - compiler/GHC/Core/Opt/OccurAnal.hs - compiler/GHC/Core/Opt/WorkWrap/Utils.hs - compiler/GHC/Core/Subst.hs - compiler/GHC/Driver/Config/Core/Lint.hs - compiler/GHC/Hs/Expr.hs - compiler/GHC/Rename/Expr.hs - compiler/GHC/Rename/Utils.hs - compiler/GHC/Tc/Gen/Expand.hs - compiler/GHC/Tc/Gen/Expr.hs - compiler/GHC/Tc/TyCl/PatSyn.hs - compiler/GHC/Tc/Types/Origin.hs - compiler/GHC/Unit/State.hs - compiler/ghc.cabal.in - libraries/base/tests/perf/all.T - libraries/ghc-boot/GHC/Unit/Database.hs - libraries/ghc-boot/ghc-boot.cabal.in - testsuite/tests/cabal/Makefile - testsuite/tests/cabal/all.T - + testsuite/tests/cabal/ghcpkg10.stdout - testsuite/tests/corelint/LintEtaExpand.stderr - testsuite/tests/corelint/T21115b.stderr - testsuite/tests/parser/should_fail/RecordDotSyntaxFail8.stderr - testsuite/tests/parser/should_fail/RecordDotSyntaxFail9.stderr - utils/ghc-pkg/Main.hs - utils/ghc-pkg/ghc-pkg.cabal.in The diff was not included because it is too large. View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/5e5b9c7364fc4d3faf79c2d829c1ef… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/5e5b9c7364fc4d3faf79c2d829c1ef… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/spj-reinstallable-base2] Wibble
by Simon Peyton Jones (＠simonpj) 19 Apr '26

19 Apr '26

Simon Peyton Jones pushed to branch wip/spj-reinstallable-base2 at Glasgow Haskell Compiler / GHC Commits: 496fc148 by Simon Peyton Jones at 2026-04-20T00:42:50+01:00 Wibble - - - - - 1 changed file: - libraries/base/src/GHC/KnownKeyNames.hs Changes: ===================================== libraries/base/src/GHC/KnownKeyNames.hs ===================================== @@ -203,6 +203,7 @@ module GHC.KnownKeyNames , Match, match , Clause, clause , Stmt, bindS, letS, noBindS, parS, recS + , Body, normalB, guardedB ) where import GHC.Internal.Base hiding( foldr ) View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/496fc148b728107133588fa126515b2… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/496fc148b728107133588fa126515b2… You're receiving this email because of your account on gitlab.haskell.org.

1 0