- ghc-commits - Haskell.org

[Git][ghc/ghc][wip/supersven/cross-testsuite] Set config.arch
by Sven Tennie (＠supersven) 31 May '26

31 May '26

Sven Tennie pushed to branch wip/supersven/cross-testsuite at Glasgow Haskell Compiler / GHC Commits: a7f1e1ff by Sven Tennie at 2026-05-31T14:34:39+02:00 Set config.arch - - - - - 1 changed file: - .gitlab/ci.sh Changes: ===================================== .gitlab/ci.sh ===================================== @@ -730,11 +730,13 @@ EOF chmod +x "${iserv_wrapper}" export EXTRA_HC_OPTS="${EXTRA_HC_OPTS:+${EXTRA_HC_OPTS} }-fexternal-interpreter -pgmi ${iserv_wrapper}" fi + local cross_arch="${CROSS_TARGET%%-*}" run_hadrian \ test \ --summary-junit=./junit.xml \ --test-have-intree-files \ --test-compiler="${test_compiler}" \ + "runtest.opts+=-e config.arch=${cross_arch}" \ "runtest.opts+=${RUNTEST_ARGS:-}" \ "runtest.opts+=--unexpected-output-dir=$TOP/unexpected-test-output" \ || fail "hadrian cross-compiled testsuite ($CROSS_TARGET)" View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/a7f1e1ffcf8b9754cd107e438f8b487… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/a7f1e1ffcf8b9754cd107e438f8b487… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/romes/hadrian-cross-stage2-rebase_SVEN_FINAL] Fix LibDir
by Sven Tennie (＠supersven) 31 May '26

31 May '26

Sven Tennie pushed to branch wip/romes/hadrian-cross-stage2-rebase_SVEN_FINAL at Glasgow Haskell Compiler / GHC Commits: b7170645 by Sven Tennie at 2026-05-31T14:28:02+02:00 Fix LibDir - - - - - 1 changed file: - hadrian/src/Rules/Generate.hs Changes: ===================================== hadrian/src/Rules/Generate.hs ===================================== @@ -271,7 +271,11 @@ generateRules = do pkgDb <- get_pkg_db libStage -- addTrailingPathSeparator needed: makeRelativeNoSysLink uses -- splitPath where "lib" and "lib/" are distinct components. - let libTopDir = addTrailingPathSeparator (root -/- stageString libStage -/- "lib") + isCrossLibDir <- crossStage stage + let libTopDir = addTrailingPathSeparator $ + if isCrossLibDir + then root -/- stageString stage' -/- "lib" + else prefix relPkgDb = makeRelativeNoSysLink libTopDir pkgDb go (generateSettings out True relPkgDb) out (prefix -/- "targets" -/- "default.target") %> \out -> go (show <$> expr (targetStage (succStage stage))) out View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/b7170645641ee26bdfe3569de358ce4… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/b7170645641ee26bdfe3569de358ce4… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/supersven/cross-testsuite] Set config.arch
by Sven Tennie (＠supersven) 31 May '26

31 May '26

Sven Tennie pushed to branch wip/supersven/cross-testsuite at Glasgow Haskell Compiler / GHC Commits: 6735b29f by Sven Tennie at 2026-05-31T11:21:15+02:00 Set config.arch - - - - - 1 changed file: - .gitlab/ci.sh Changes: ===================================== .gitlab/ci.sh ===================================== @@ -730,11 +730,13 @@ EOF chmod +x "${iserv_wrapper}" export EXTRA_HC_OPTS="${EXTRA_HC_OPTS:+${EXTRA_HC_OPTS} }-fexternal-interpreter -pgmi ${iserv_wrapper}" fi + local cross_arch="${CROSS_TARGET%%-*}" run_hadrian \ test \ --summary-junit=./junit.xml \ --test-have-intree-files \ --test-compiler="${test_compiler}" \ + --config "config.arch='${cross_arch}'" \ "runtest.opts+=${RUNTEST_ARGS:-}" \ "runtest.opts+=--unexpected-output-dir=$TOP/unexpected-test-output" \ || fail "hadrian cross-compiled testsuite ($CROSS_TARGET)" View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/6735b29fe4d2fa31d7d29f455f2924e… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/6735b29fe4d2fa31d7d29f455f2924e… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/romes/hadrian-cross-stage2-rebase_SVEN_FINAL] 19 commits: Hadrian: create a ghc-internal .def file per ghc-internal dll
by Sven Tennie (＠supersven) 31 May '26

31 May '26

Sven Tennie pushed to branch wip/romes/hadrian-cross-stage2-rebase_SVEN_FINAL at Glasgow Haskell Compiler / GHC Commits: d603477f by David Eichmann at 2026-05-29T13:17:12-04:00 Hadrian: create a ghc-internal .def file per ghc-internal dll The .def file generated from rts/win32/libHSghc-internal.def.in contains the name of the ghc-internal dll. The correct dll name differs based on if the dll is inplace/final and if using the Dynamic way. Previously, this was not accounted for and inconsistent dlls names where used. That led to failure when loading dlls at runtime in experiments with windows dynamic linking. - - - - - 1fc21753 by Sylvain Henry at 2026-05-29T13:18:14-04:00 ghc-bignum: copy backend interface haddocks to Native backend (#27305) The haddock comments documenting the BigNat backend interface (function contracts, expected MutableWordArray# sizes, return-value semantics, etc.) were attached to the FFI backend module. Copy them to the Native backend so they remain in tree once the FFI backend is removed. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply(a)anthropic.com> - - - - - 717059df by Sylvain Henry at 2026-05-29T13:18:14-04:00 ghc-bignum: remove FFI backend (#27305) The FFI backend of ghc-bignum (now part of ghc-internal) had no known users and is easy to recreate by relinking ghc-internal with a custom backend. Remove the backend module, the bignum-ffi cabal flag, and the ffi option from Hadrian's --bignum selector. The backend interface documentation now lives in the Native backend module. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply(a)anthropic.com> - - - - - 4bb3b1d8 by Sylvain Henry at 2026-05-29T13:18:14-04:00 ghc-bignum: remove Check backend (#27305) The Check backend of ghc-bignum (now part of ghc-internal) compared the selected backend's output against the Native backend for validation. It had no known users. Remove the backend module, the bignum-check cabal flag, the bignumCheck Hadrian flavour field, and the check- prefix in Hadrian's --bignum selector. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply(a)anthropic.com> - - - - - 6b3044a0 by David Eichmann at 2026-05-30T11:58:48-04:00 Add code comments to allocator code - - - - - f4e04210 by Matthew Pickering at 2026-05-30T11:59:34-04:00 hadrian: Refactor system-cxx-std-lib rules I noticed a few things wrong with the hadrian rules for `system-cxx-std-lib` rules. * For `text` there is an ad-hoc check to depend on `system-cxx-std-lib` outside of `configurePackage`. * The `system-cxx-std-lib` dependency is not read from cabal files. * Recache is not called on the packge database after the `.conf` file is generated, a more natural place for this rule is `registerRules`. Treating this uniformly like other packages is complicated by it not having any source code or a cabal file. However we can do a bit better by reporting the dependency firstly in `PackageData` and then needing the `.conf` file in the same place as every other package in `configurePackage`. This commit increases the `shakeVersion`, to provide backwards compatibility to previous builds with different PackageData. Fixes #25303 Co-authored-by: Sven Tennie <sven.tennie(a)gmail.com> - - - - - fe2b480f by Matthew Pickering at 2026-05-30T19:50:58+02:00 Add missing req_interp modifier to T18441fail3 and T18441fail19 These tests require the interpreter but they were failing in a different way with the javascript backend because the interpreter was disabled and stderr is ignored by the test. - - - - - c745418e by Matthew Pickering at 2026-05-30T19:50:58+02:00 hadrian: Fill in more of the default.host toolchain file When you are building a cross compiler this file will be used to build stage1 and it's libraries, so we need enough information here to work accurately. There is still more work to be done (see for example, word size is still fixed). - - - - - 38a791a5 by Matthew Pickering at 2026-05-30T19:50:58+02:00 hadrian: Disable docs when cross compiling Before there were a variety of ad-hoc places where doc building was disabled when cross compiling. * Some CI jobs sets --docs=none in gen_ci.hs * Some CI jobs set --docs=none in .gitlab/ci.sh * There was some logic in hadrian to not need the ["docs"] target when making a bindist. Now the situation is simple: * If you are cross compiling then defaultDocsTargets is empty by default. In theory, there is no reason why we can't build documentation for cross compiler bindists, but this is left to future work to generalise the documentation building rules to allow this (#24289) - - - - - bcd91432 by Matthew Pickering at 2026-05-30T23:03:20+02:00 hadrian: Build stage 2 cross compilers * Most of hadrian is abstracted over the stage in order to remove the assumption that the target of all stages is the same platform. This allows the RTS to be built for two different targets for example. * Abstracts the bindist creation logic to allow building either normal or cross bindists. Normal bindists use stage 1 libraries and a stage 2 compiler. Cross bindists use stage 2 libararies and a stage 2 compiler. * hadrian: Make binary-dist-dir the default build target. This allows us to have the logic in one place about which libraries/stages to build with cross compilers. Fixes #24192 New hadrian target: * `binary-dist-dir-cross`: Build a cross compiler bindist (compiler = stage 1, libraries = stage 2) This commit also contains various changes to make stage2 compilers feasible. ------------------------- Metric Decrease: ManyAlternatives MultiComponentModulesRecomp MultiLayerModulesRecomp T10421 T12425 T12707 T13035 T13379 T15703 T16577 T18698a T18698b T18923 T1969 T21839c T3294 T4801 T5030 T5321Fun T5642 T783 T9198 T9872d T9961 parsing001 T5321FD T6048 T12227 T18140 T18282 T9233 T5631 T9630 ------------------------- Co-authored-by: Sven Tennie <sven.tennie(a)gmail.com> Fix rebase: settings-use-distro-mingw is now staged - - - - - 2fa22d0f by Matthew Pickering at 2026-05-30T23:03:20+02:00 ci: Test cross bindists We remove the special logic for testing in-tree cross compilers and instead test cross compiler bindists, like we do for all other platforms. - - - - - 5c03d635 by Matthew Pickering at 2026-05-30T23:03:20+02:00 ci: Introduce CROSS_STAGE variable In preparation for building and testing stage3 bindists we introduce the CROSS_STAGE variable which is used by a CI job to determine what kind of bindist the CI job should produce. At the moment we are only using CROSS_STAGE=2 but in the future we will have some jobs which set CROSS_STAGE=3 to produce native bindists for a target, but produced by a cross compiler, which can be tested on by another CI job on the native platform. CROSS_STAGE=2: Build a normal cross compiler bindist CROSS_STAGE=3: Build a stage 3 bindist, one which is a native compiler and library for the target - - - - - b4b00940 by Sven Tennie at 2026-05-30T23:16:13+02:00 ghc: Distinguish between having an interpreter and having an internal one Otherwise, we fail with warnings when compiling tools. Actually, these are related but different things: - ghc can run an interpreter (either internal or external) - ghc is compiled with an internal interpreter - - - - - 06ec04f2 by Sven Tennie at 2026-05-30T23:16:14+02:00 ci: Increase timeout for emulators Test runs with emulators naturally take longer than on native machines. Generate jobs.yml - - - - - d22086c7 by Matthew Pickering at 2026-05-30T23:16:14+02:00 ci: Javascript don't set CROSS_EMULATOR There is no CROSS_EMULATOR needed to run javascript binaries, so we don't set the CROSS_EMULATOR to some dummy value. - - - - - 8a1bdec6 by Sven Tennie at 2026-05-30T23:16:14+02:00 Javascript skip T23697 See #22355 about how HSC2HS and the Javascript target don't play well together. - - - - - d500a84a by Sven Tennie at 2026-05-30T23:16:14+02:00 Mark T24602 as fragile It was skipped before (due to CROSS_EMULATOR being set, which changed for JS), so we don't make things worse by marking it as fragile. - - - - - 78e974cf by Sven Tennie at 2026-05-30T23:16:14+02:00 Fix libdir - - - - - ccb507db by Sven Tennie at 2026-05-30T23:16:14+02:00 Fix cross LibDir - - - - - 82 changed files: - .gitlab/ci.sh - .gitlab/generate-ci/gen_ci.hs - .gitlab/jobs.yaml - + changelog.d/hadrian-system-cxx-std-lib-25303 - + changelog.d/remove-bignum-check-backend - + changelog.d/remove-bignum-ffi-backend - configure.ac - distrib/configure.ac.in - hadrian/README.md - hadrian/bindist/config.mk.in - hadrian/cfg/default.host.target.in - + hadrian/cfg/system.config.host.in - hadrian/cfg/system.config.in - + hadrian/cfg/system.config.target.in - hadrian/doc/user-settings.md - hadrian/hadrian.cabal - hadrian/src/Base.hs - + hadrian/src/BindistConfig.hs - hadrian/src/Builder.hs - hadrian/src/CommandLine.hs - hadrian/src/Context.hs - hadrian/src/Expression.hs - hadrian/src/Flavour.hs - hadrian/src/Flavour/Type.hs - hadrian/src/Hadrian/Builder.hs - hadrian/src/Hadrian/Haskell/Cabal/Parse.hs - hadrian/src/Hadrian/Haskell/Cabal/Type.hs - hadrian/src/Hadrian/Haskell/Hash.hs - hadrian/src/Hadrian/Oracles/TextFile.hs - hadrian/src/Main.hs - hadrian/src/Oracles/Flag.hs - hadrian/src/Oracles/Flavour.hs - hadrian/src/Oracles/Setting.hs - hadrian/src/Oracles/TestSettings.hs - hadrian/src/Packages.hs - hadrian/src/Rules.hs - hadrian/src/Rules/BinaryDist.hs - hadrian/src/Rules/CabalReinstall.hs - hadrian/src/Rules/Changelog.hs - hadrian/src/Rules/Compile.hs - hadrian/src/Rules/Documentation.hs - hadrian/src/Rules/Generate.hs - hadrian/src/Rules/Gmp.hs - hadrian/src/Rules/Library.hs - hadrian/src/Rules/Program.hs - hadrian/src/Rules/Register.hs - hadrian/src/Rules/Rts.hs - hadrian/src/Rules/Test.hs - hadrian/src/Settings.hs - hadrian/src/Settings/Builders/Cabal.hs - hadrian/src/Settings/Builders/Common.hs - hadrian/src/Settings/Builders/Configure.hs - hadrian/src/Settings/Builders/DeriveConstants.hs - hadrian/src/Settings/Builders/Ghc.hs - hadrian/src/Settings/Builders/Hsc2Hs.hs - hadrian/src/Settings/Builders/RunTest.hs - hadrian/src/Settings/Builders/SplitSections.hs - hadrian/src/Settings/Default.hs - hadrian/src/Settings/Flavours/GhcInGhci.hs - hadrian/src/Settings/Flavours/Performance.hs - hadrian/src/Settings/Flavours/QuickCross.hs - hadrian/src/Settings/Flavours/Validate.hs - hadrian/src/Settings/Packages.hs - hadrian/src/Settings/Program.hs - hadrian/src/Settings/Warnings.hs - libraries/base/tests/all.T - libraries/ghc-bignum/ghc-bignum.cabal - libraries/ghc-internal/bignum-backend.rst - libraries/ghc-internal/ghc-internal.cabal.in - libraries/ghc-internal/src/GHC/Internal/Bignum/Backend.hs - − libraries/ghc-internal/src/GHC/Internal/Bignum/Backend/Check.hs - − libraries/ghc-internal/src/GHC/Internal/Bignum/Backend/FFI.hs - libraries/ghc-internal/src/GHC/Internal/Bignum/Backend/Native.hs - − libraries/ghc-internal/src/GHC/Internal/Bignum/Backend/Selected.hs - m4/fp_find_nm.m4 - m4/prep_target_file.m4 - rts/sm/BlockAlloc.c - rts/sm/MBlock.c - rts/win32/libHSghc-internal.def.in - testsuite/ghc-config/ghc-config.hs - testsuite/tests/ghc-e/should_fail/all.T - testsuite/tests/javascript/closure/all.T The diff was not included because it is too large. View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/a3b31d8ec84bbaf0c3b318579c2d0b… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/a3b31d8ec84bbaf0c3b318579c2d0b… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/supersven/cross-testsuite] Create iserv_wrapper, execute it with Qemu
by Sven Tennie (＠supersven) 30 May '26

30 May '26

Sven Tennie pushed to branch wip/supersven/cross-testsuite at Glasgow Haskell Compiler / GHC Commits: fd9c75a0 by Sven Tennie at 2026-05-30T23:49:55+02:00 Create iserv_wrapper, execute it with Qemu - - - - - 1 changed file: - .gitlab/ci.sh Changes: ===================================== .gitlab/ci.sh ===================================== @@ -709,7 +709,27 @@ function test_hadrian() { # > main = putStrLn "hello world" run diff -w expected actual - EXTRA_HC_OPTS="${EXTRA_HC_OPTS:+${EXTRA_HC_OPTS} }-fexternal-interpreter" + if [[ "${CROSS_EMULATOR:-}" == qemu-* ]]; then + cat > "${TOP}/tmp/iserv_main.hs" <<'EOF' +import GHCi.Server (defaultServer) +main :: IO () +main = defaultServer +EOF + local iserv_bin="${TOP}/tmp/ghc-iserv-cross" + run "${test_compiler}" \ + -package ghci \ + -threaded \ + -optl-Wl,--export-dynamic \ + "${TOP}/tmp/iserv_main.hs" \ + -o "${iserv_bin}" + local iserv_wrapper="${TOP}/tmp/iserv-wrapper.sh" + cat > "${iserv_wrapper}" <<EOF +#!/usr/bin/env bash +exec ${CROSS_EMULATOR} "${iserv_bin}" "\$@" +EOF + chmod +x "${iserv_wrapper}" + export EXTRA_HC_OPTS="${EXTRA_HC_OPTS:+${EXTRA_HC_OPTS} }-fexternal-interpreter -pgmi ${iserv_wrapper}" + fi run_hadrian \ test \ --summary-junit=./junit.xml \ View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/fd9c75a0de21f0061b2c72d724e6fa7… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/fd9c75a0de21f0061b2c72d724e6fa7… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/supersven/cross-testsuite] 2 commits: Add -fexternal-interpreter
by Sven Tennie (＠supersven) 30 May '26

30 May '26

Sven Tennie pushed to branch wip/supersven/cross-testsuite at Glasgow Haskell Compiler / GHC Commits: 48a3f629 by Sven Tennie at 2026-05-30T19:45:00+02:00 Add -fexternal-interpreter - - - - - 61b360c0 by Sven Tennie at 2026-05-30T19:45:15+02:00 Mark tests as broken - - - - - 2 changed files: - .gitlab/ci.sh - testsuite/tests/numeric/should_run/all.T Changes: ===================================== .gitlab/ci.sh ===================================== @@ -709,6 +709,7 @@ function test_hadrian() { # > main = putStrLn "hello world" run diff -w expected actual + EXTRA_HC_OPTS="${EXTRA_HC_OPTS:+${EXTRA_HC_OPTS} }-fexternal-interpreter" run_hadrian \ test \ --summary-junit=./junit.xml \ ===================================== testsuite/tests/numeric/should_run/all.T ===================================== @@ -11,7 +11,7 @@ test('arith001', normal, compile_and_run, ['']) test('arith002', normal, compile_and_run, ['']) test('arith003', normal, compile_and_run, ['']) test('arith004', normal, compile_and_run, ['']) -test('arith005', normal, compile_and_run, ['']) +test('arith005', [when(arch('riscv64'), expect_broken(27303))], compile_and_run, ['']) test('arith006', normal, compile_and_run, ['']) test('arith007', normal, compile_and_run, ['']) @@ -45,7 +45,7 @@ test('arith018', normal, compile_and_run, ['']) test('arith019', normal, compile_and_run, ['']) test('expfloat', normal, compile_and_run, ['']) -test('FloatFnInverses', normal, compile_and_run, ['']) +test('FloatFnInverses', [when(arch('riscv64'), expect_broken(27306))], compile_and_run, ['']) test('T1603', skip, compile_and_run, ['']) test('T3676', expect_broken(3676), compile_and_run, ['']) View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/b8c534d9d85f9d7f3b6b17bf26b253… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/b8c534d9d85f9d7f3b6b17bf26b253… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][master] hadrian: Refactor system-cxx-std-lib rules
by Marge Bot (＠marge-bot) 30 May '26

30 May '26

Marge Bot pushed to branch master at Glasgow Haskell Compiler / GHC Commits: f4e04210 by Matthew Pickering at 2026-05-30T11:59:34-04:00 hadrian: Refactor system-cxx-std-lib rules I noticed a few things wrong with the hadrian rules for `system-cxx-std-lib` rules. * For `text` there is an ad-hoc check to depend on `system-cxx-std-lib` outside of `configurePackage`. * The `system-cxx-std-lib` dependency is not read from cabal files. * Recache is not called on the packge database after the `.conf` file is generated, a more natural place for this rule is `registerRules`. Treating this uniformly like other packages is complicated by it not having any source code or a cabal file. However we can do a bit better by reporting the dependency firstly in `PackageData` and then needing the `.conf` file in the same place as every other package in `configurePackage`. This commit increases the `shakeVersion`, to provide backwards compatibility to previous builds with different PackageData. Fixes #25303 Co-authored-by: Sven Tennie <sven.tennie(a)gmail.com> - - - - - 6 changed files: - + changelog.d/hadrian-system-cxx-std-lib-25303 - hadrian/src/Hadrian/Haskell/Cabal/Parse.hs - hadrian/src/Hadrian/Haskell/Cabal/Type.hs - hadrian/src/Main.hs - hadrian/src/Rules/Generate.hs - hadrian/src/Rules/Register.hs Changes: ===================================== changelog.d/hadrian-system-cxx-std-lib-25303 ===================================== @@ -0,0 +1,22 @@ +section: packaging +synopsis: Fix Hadrian rules for system-cxx-std-lib package dependency +issues: #25303 +mrs: !16013 +description: { + Hadrian's handling of the `system-cxx-std-lib` virtual package has been + fixed and made more uniform. + + Previously, `text` had an ad-hoc rule outside of `configurePackage` to + declare a dependency on `system-cxx-std-lib`, the dependency was not + discovered from cabal files, and the package database was not recached after + the `.conf` file was generated. + + The dependency is now read from cabal files via a new + `dependsOnSystemCxxStdLib` field in `PackageData`, and the `.conf` file + is needed inside `configurePackage` alongside all other package + dependencies, consistent with how every other package is handled. + + `shakeVersion` has been bumped to ensure existing build databases are + invalidated when upgrading, preventing binary deserialisation errors due to + the changed `PackageData` type. +} ===================================== hadrian/src/Hadrian/Haskell/Cabal/Parse.hs ===================================== @@ -81,10 +81,11 @@ parsePackageData pkg = do sorted = sort [ C.unPackageName p | C.Dependency p _ _ <- allDeps ] deps = nubOrd sorted \\ [name] depPkgs = mapMaybe findPackageByName deps + cxxStdLib = elem "system-cxx-std-lib" deps return $ PackageData name version (C.fromShortText (C.synopsis pd)) (C.fromShortText (C.description pd)) - depPkgs gpd + depPkgs cxxStdLib gpd where -- Collect an overapproximation of dependencies by ignoring conditionals collectDeps :: Maybe (C.CondTree v [C.Dependency] a) -> [C.Dependency] @@ -138,7 +139,9 @@ configurePackage :: Context -> Action () configurePackage context@Context {..} = do putProgressInfo $ "| Configure package " ++ quote (pkgName package) gpd <- pkgGenericDescription package - depPkgs <- packageDependencies <$> readPackageData package + pd <- readPackageData package + let depPkgs = packageDependencies pd + needSystemCxxStdLib = dependsOnSystemCxxStdLib pd -- Stage packages are those we have in this stage. stagePkgs <- stagePackages stage @@ -157,7 +160,12 @@ configurePackage context@Context {..} = do -- We'll need those packages in our package database. deps <- sequence [ pkgConfFile (context { package = pkg, iplace = forceBaseAfterGhcInternal pkg }) | pkg <- depPkgs, pkg `elem` stagePkgs ] - need $ extraPreConfigureDeps ++ deps + -- system-cxx-std-lib is magic.. it doesn't have a cabal file or source code, so we have + -- to treat it specially as `pkgConfFile` uses `readPackageData` to compute the version. + systemCxxStdLib <- sequence [ systemCxxStdLibConfPath (PackageDbLoc stage iplace) | needSystemCxxStdLib ] + need $ extraPreConfigureDeps + ++ deps + ++ systemCxxStdLib -- Figure out what hooks we need. let configureFile = replaceFileName (pkgCabalFile package) "configure" ===================================== hadrian/src/Hadrian/Haskell/Cabal/Type.hs ===================================== @@ -30,6 +30,7 @@ data PackageData = PackageData , synopsis :: String , description :: String , packageDependencies :: [Package] + , dependsOnSystemCxxStdLib :: Bool , genericPackageDescription :: GenericPackageDescription } deriving (Eq, Generic, Show) ===================================== hadrian/src/Main.hs ===================================== @@ -63,7 +63,13 @@ main = do shakeColor <- shouldUseColor let options :: ShakeOptions options = shakeOptions - { shakeChange = ChangeModtimeAndDigest + { -- Bump shakeVersion whenever a type stored in the Shake oracle + -- changes its Binary representation (e.g. fields added/removed + -- from PackageData or other oracle value types). This forces + -- Shake to wipe the stale database instead of crashing on + -- deserialisation. + shakeVersion = "2" + , shakeChange = ChangeModtimeAndDigest , shakeFiles = buildRoot -/- Base.shakeFilesDir , shakeProgress = Progress.hadrianProgress cwd , shakeRebuild = rebuild ===================================== hadrian/src/Rules/Generate.hs ===================================== @@ -242,9 +242,6 @@ copyRules = do prefix -/- "html/**" <~ return "utils/haddock/haddock-api/resources" prefix -/- "latex/**" <~ return "utils/haddock/haddock-api/resources" - forM_ [Inplace, Final] $ \iplace -> - root -/- relativePackageDbPath (PackageDbLoc stage iplace) -/- systemCxxStdLibConf %> \file -> do - copyFile ("mk" -/- "system-cxx-std-lib-1.0.conf") file generateRules :: Rules () generateRules = do ===================================== hadrian/src/Rules/Register.hs ===================================== @@ -6,7 +6,6 @@ module Rules.Register ( import Base import Context -import Flavour import Oracles.Setting import Hadrian.BuildPath import Hadrian.Expression @@ -48,14 +47,6 @@ configurePackageRules = do isGmp <- (== "gmp") <$> interpretInContext ctx getBignumBackend when isGmp $ need [buildP -/- "include/ghc-gmp.h"] - when (pkg == text) $ do - simdutf <- textWithSIMDUTF <$> flavour - when simdutf $ do - -- This is required, otherwise you get Error: hadrian: - -- Encountered missing or private dependencies: - -- system-cxx-std-lib ==1.0 - cxxStdLib <- systemCxxStdLibConfPath $ PackageDbLoc stage Inplace - need [cxxStdLib] Cabal.configurePackage ctx root -/- "**/autogen/cabal_macros.h" %> \out -> do @@ -105,6 +96,12 @@ registerPackageRules rs stage iplace = do target (Context stage compiler vanilla iplace) (GhcPkg Recache stage) [] [] writeFileLines stamp [] + -- Special rule for registering system-cxx-std-lib + root -/- relativePackageDbPath (PackageDbLoc stage iplace) -/- systemCxxStdLibConf %> \file -> do + copyFile ("mk" -/- "system-cxx-std-lib-1.0.conf") file + buildWithResources rs $ + target (Context stage compiler vanilla iplace) (GhcPkg Recache stage) [] [] + -- Register a package. root -/- relativePackageDbPath (PackageDbLoc stage iplace) -/- "*.conf" %> \conf -> do historyDisable View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/f4e04210f8233e912d8c077e95442db… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/f4e04210f8233e912d8c077e95442db… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][master] Add code comments to allocator code
by Marge Bot (＠marge-bot) 30 May '26

30 May '26

Marge Bot pushed to branch master at Glasgow Haskell Compiler / GHC Commits: 6b3044a0 by David Eichmann at 2026-05-30T11:58:48-04:00 Add code comments to allocator code - - - - - 2 changed files: - rts/sm/BlockAlloc.c - rts/sm/MBlock.c Changes: ===================================== rts/sm/BlockAlloc.c ===================================== @@ -549,6 +549,8 @@ allocGroupOnNode (uint32_t node, W_ n) ln++; } + // If no free blocks exist then allocate a new megablock and keep just a + // chunk of it. if (ln == NUM_FREE_LISTS) { #if 0 /* useful for debugging fragmentation */ if ((W_)mblocks_allocated * BLOCKS_PER_MBLOCK * BLOCK_SIZE_W @@ -1020,6 +1022,9 @@ freeGroup(bdescr *p) } // coalesce backwards + // Note that p is not a megablock/megagroup, so there are still live blocks on + // this megablock. Hence we can't coalesce backwards past the first block + // descriptor of this megablock. if (p != FIRST_BDESCR(MBLOCK_ROUND_DOWN(p))) { bdescr *prev; ===================================== rts/sm/MBlock.c ===================================== @@ -95,6 +95,10 @@ typedef struct free_list { } free_list; static free_list *free_list_head; + +// The address in `mblock_address_space` just after the highest MBlock. This +// will be the address of the next new MBlock committed when the free list is +// empty. static W_ mblock_high_watermark; /* * it is quite important that these are in the same cache line as they @@ -103,22 +107,47 @@ static W_ mblock_high_watermark; */ struct mblock_address_range mblock_address_space = { 0, 0, {} }; +// Search for the first committed MBlock at or after startingAt. +// +// start_iter [in/out]: The free_list or a subset of it. Must contain all +// entries for MBlocks at or after startingAt. On return this is set to the +// free_list entry just after the returned MBlock. If no MBlock was found, This +// is set to NULL (the search will have reached the end of the free list). +// +// startingAt [in]: address from which to start searching. This must be to the +// start of an MBlock. +// +// return: The address of the committed MBlock. NULL if no committed MBLock was +// found. +// static void *getAllocatedMBlock(free_list **start_iter, W_ startingAt) { + // We simultaneously traverse the free list and the mblock_address_space. free_list *iter; W_ p = startingAt; for (iter = *start_iter; iter != NULL; iter = iter->next) { + // The current free MBlock, `iter`, is past the current MBlock, `p`. + // This means that `p` is committed (if it was free, then we would have + // found it on the free list). Stop searching. if (p < iter->address) break; + // Note that if `p > iter->address`, then we don't bump `p`. This just + // means we are skipping entries in the free list that correspond to + // MBlocks before `startingAt`. if (p == iter->address) + // Move to the next MBlock. The MBlock may be committed, + // uncommitted, or even past mblock_high_watermark. p += iter->size; } + // Output current free list entry. *start_iter = iter; + // If we reached mblock_high_watermark, then we didn't find any committed + // MBlocks. if (p >= mblock_high_watermark) return NULL; @@ -152,6 +181,11 @@ void * getNextMBlock(void **state STG_UNUSED, void *mblock) return getAllocatedMBlock(casted_state, (W_)mblock + MBLOCK_SIZE); } +// Used to implement getCommittedMBlocks. Search the free list for n contiguous +// free MBlocks, and commit those MBlocks, updating the free_list. Returns the +// address of the start of those MBlocks. Returns NULL if no n contiguous +// MBlocks were found in the free list. Unlike getFreshMBlocks, this doesn't +// attempt to allocate new MBlocks past mblock_high_watermark. static void *getReusableMBlocks(uint32_t n) { struct free_list *iter; @@ -163,7 +197,10 @@ static void *getReusableMBlocks(uint32_t n) if (iter->size < size) continue; + // We've found a large enough group of MBlocks. addr = (void*)iter->address; + + // Update the free list. iter->address += size; iter->size -= size; if (iter->size == 0) { @@ -190,6 +227,9 @@ static void *getReusableMBlocks(uint32_t n) return NULL; } +// Used to implement getCommittedMBlocks. Commit n new MBlocks starting at +// mblock_high_watermark. May exit with an out of memory error if we've passed +// mblock_address_space.end (this is unlikely). static void *getFreshMBlocks(uint32_t n) { W_ size = MBLOCK_SIZE * (W_)n; @@ -207,6 +247,8 @@ static void *getFreshMBlocks(uint32_t n) return addr; } +// Commit n new MBlocks. Tries to reuse freed MBlocks, else commits new +// MBlock(s) at mblock_high_watermark. static void *getCommittedMBlocks(uint32_t n) { void *p; @@ -220,6 +262,11 @@ static void *getCommittedMBlocks(uint32_t n) return p; } +// Decommit n contiguous MBlocks starting at the given address. +// +// addr [in]: address of the start of the n MBlocks (in mblock_address_space). +// +// n [in]: number of contiguous MBlocks to decommit. static void decommitMBlocks(char *addr, uint32_t n) { struct free_list *iter, *prev; @@ -228,17 +275,24 @@ static void decommitMBlocks(char *addr, uint32_t n) osDecommitMemory(addr, size); + // Update the free list. prev = NULL; for (iter = free_list_head; iter != NULL; iter = iter->next) { prev = iter; + // iter is still entirely behind and not contiguous with the MBlocks so + // continue traversing free_list. if (iter->address + iter->size < address) continue; + // The MBlocks are after and contiguous to iter. Simply modify the + // current entry to include n more MBlocks and possibly coalesce. if (iter->address + iter->size == address) { iter->size += size; + // If the current free_list entry now reaches mblock_high_watermark, + // remove the entry and decrement mblock_high_watermark. if (address + size == mblock_high_watermark) { mblock_high_watermark -= iter->size; if (iter->prev) { @@ -251,6 +305,8 @@ static void decommitMBlocks(char *addr, uint32_t n) return; } + // If the current free_list entry now reaches the next free_list + // entry, coalesce them. if (iter->next && iter->next->address == iter->address + iter->size) { struct free_list *next; @@ -269,6 +325,8 @@ static void decommitMBlocks(char *addr, uint32_t n) stgFree(next); } return; + + // The MBlocks are before and contiguous to iter. } else if (address + size == iter->address) { iter->address = address; iter->size += size; @@ -280,6 +338,9 @@ static void decommitMBlocks(char *addr, uint32_t n) ASSERT(iter->prev->address + iter->prev->size < iter->address); } return; + + // The MBlocks are before and not contiguous to iter. Insert a new entry + // just before iter. } else { struct free_list *new_iter; @@ -311,6 +372,7 @@ static void decommitMBlocks(char *addr, uint32_t n) if (address + size == mblock_high_watermark) { mblock_high_watermark -= size; } else { + // Add a new entry to the end of the free list. struct free_list *new_iter; new_iter = stgMallocBytes(sizeof(struct free_list), "freeMBlocks"); View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/6b3044a038a94e323478eac4278632c… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/6b3044a038a94e323478eac4278632c… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/jeltsch/textual-bytecode-output] A bit of wobbling
by Wolfgang Jeltsch (＠jeltsch) 30 May '26

30 May '26

Wolfgang Jeltsch pushed to branch wip/jeltsch/textual-bytecode-output at Glasgow Haskell Compiler / GHC Commits: 88c0c200 by Wolfgang Jeltsch at 2026-05-30T18:52:20+03:00 A bit of wobbling - - - - - 1 changed file: - compiler/GHC/ByteCode/Show.hs Changes: ===================================== compiler/GHC/ByteCode/Show.hs ===================================== @@ -37,7 +37,12 @@ import GHC.Utils.Outputable withPprStyle, ppr ) -import GHC.ByteCode.Types (UnlinkedBCO, ByteCodeHpcInfo, CompiledByteCode (..)) +import GHC.ByteCode.Types + ( + UnlinkedBCO (UnlinkedBCO, UnlinkedStaticCon), + ByteCodeHpcInfo, + CompiledByteCode (..) + ) import GHC.ByteCode.Breakpoints (InternalModBreaks) import GHC.ByteCode.Binary (OnDiskModuleByteCode (..)) import GHC.ByteCode.Serialize (readOnDiskModuleByteCode) @@ -78,17 +83,44 @@ pprCompiledByteCode CompiledByteCode {..} pprBreakPoints bc_breaks, pprStaticPointerTableEntries bc_spt_entries, pprHPCInfo bc_hpc_info - ] + ] -- | […] pprByteCodeObjects :: [UnlinkedBCO] -> SDoc -pprByteCodeObjects = multiLineEntry (text "bytecode objects") . - vcatOrNone . +pprByteCodeObjects = multiLineEntry (text "bytecode objects:") . + vcatOrNone . map pprByteCodeObject -- | […] pprByteCodeObject :: UnlinkedBCO -> SDoc -pprByteCodeObject = pprByteCodeObject +pprByteCodeObject UnlinkedBCO {{-..-}} + = multiLineEntry (text "ordinary bytecode object:") $ + vcat [ + {- + unlinkedBCOName :: !Name, + unlinkedBCOArity :: {-# UNPACK #-} !Int, + unlinkedBCOInstrs :: !(BCOByteArray Word16), -- insns + unlinkedBCOBitmap :: !(BCOByteArray Word), -- bitmap + unlinkedBCOLits :: !(FlatBag BCONPtr), -- non-ptrs + unlinkedBCOPtrs :: !(FlatBag BCOPtr) -- ptrs + -} + ] +pprByteCodeObject UnlinkedStaticCon {{-..-}} + = multiLineEntry (text "static constructor object:") $ + vcat [ + {- + unlinkedStaticConName :: !Name, + -- ^ The name to which this static constructor is bound, not to be + -- confused with the name of the static constructor itself + -- ('unlinkedStaticConDataConName') + unlinkedStaticConDataConName :: !Name, + unlinkedStaticConLits :: !(FlatBag BCONPtr), + -- ^ non-ptrs full words, where sub-word literals have already been + -- packed into full words as needed + unlinkedStaticConPtrs :: !(FlatBag BCOPtr), -- ptrs + unlinkedStaticConIsUnlifted :: !Bool + -} + ] -- | […] pprDataConstructorInfoTables :: [(Name, ConInfoTable)] -> SDoc View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/88c0c200c4f6dc263681ea275477c92… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/88c0c200c4f6dc263681ea275477c92… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/jeltsch/textual-bytecode-output] 2 commits: Improvement of code structure
by Wolfgang Jeltsch (＠jeltsch) 30 May '26

30 May '26

Wolfgang Jeltsch pushed to branch wip/jeltsch/textual-bytecode-output at Glasgow Haskell Compiler / GHC Commits: a70a5cd8 by Wolfgang Jeltsch at 2026-05-30T18:29:29+03:00 Improvement of code structure - - - - - 8bc75cd9 by Wolfgang Jeltsch at 2026-05-30T18:48:35+03:00 A bit of wobbling - - - - - 1 changed file: - compiler/GHC/ByteCode/Show.hs Changes: ===================================== compiler/GHC/ByteCode/Show.hs ===================================== @@ -10,14 +10,15 @@ import Data.Function (($), (&), (.)) import Data.Bool (otherwise) import Data.Int (Int) import Data.Word (Word8) -import Data.Maybe (maybe) +import Data.Maybe (Maybe, maybe) import Data.List (map, zipWith) import Data.String (String) import Data.ByteString (ByteString, null, unpack) import Numeric (showHex) import System.IO (IO, FilePath) -import GHC.Data.Strict qualified as Strict (maybe) +import GHC.Data.Strict qualified as Strict (Maybe, maybe) import GHC.Data.FlatBag (elemsFlatBag) +import GHC.Fingerprint (Fingerprint) import GHC.Types.SrcLoc (noSrcSpan) import GHC.Types.Name (Name) import GHC.Types.SptEntry (SptEntry) @@ -57,74 +58,111 @@ pprOnDiskModuleByteCode :: OnDiskModuleByteCode -> SDoc pprOnDiskModuleByteCode OnDiskModuleByteCode {..} = multiLineEntry (text "module" <+> ppr odgbc_module) $ vcat [ - singleLineEntry - (text "hash:") - (ppr odgbc_hash), - multiLineEntry - (text "compiled bytecode:") - (pprCompiledByteCode odgbc_compiled_byte_code), - multiLineEntry - (text "contents of object files:") - (pprObjectFileContents odgbc_foreign) + pprHash odgbc_hash, + pprCompiledByteCode odgbc_compiled_byte_code, + pprObjectFileContents odgbc_foreign ] +-- | […] +pprHash :: Fingerprint -> SDoc +pprHash hash = singleLineEntry (text "hash:") (ppr hash) + -- | […] pprCompiledByteCode :: CompiledByteCode -> SDoc pprCompiledByteCode CompiledByteCode {..} - = vcat - [ - multiLineEntry - (text "bytecode objects") - (pprByteCodeObjects (elemsFlatBag bc_bcos)), - multiLineEntry - (text "data constructor info tables:") - (pprDataConstructorInfoTables bc_itbls), - multiLineEntry - (text "top-level strings:") - (pprTopLevelStrings bc_strs), - multiLineEntry - (text "break points:") - (maybe (text "<none>") pprInternalBreakPoints bc_breaks), - multiLineEntry - (text "static-pointer table entries:") - (pprStaticPointerTableEntries bc_spt_entries), - multiLineEntry - (text "HPC information:") - (Strict.maybe (text "<none>") pprHPCInfo bc_hpc_info) - ] + = multiLineEntry (text "compiled bytecode:") $ + vcat [ + pprByteCodeObjects (elemsFlatBag bc_bcos), + pprDataConstructorInfoTables bc_itbls, + pprTopLevelStrings bc_strs, + pprBreakPoints bc_breaks, + pprStaticPointerTableEntries bc_spt_entries, + pprHPCInfo bc_hpc_info + ] -- | […] pprByteCodeObjects :: [UnlinkedBCO] -> SDoc -pprByteCodeObjects = vcatOrNone . map pprByteCodeObject +pprByteCodeObjects = multiLineEntry (text "bytecode objects:") . + vcatOrNone . + map pprByteCodeObject -- | […] pprByteCodeObject :: UnlinkedBCO -> SDoc -pprByteCodeObject {- UnlinkedBCO {..} -} - = pprByteCodeObject +pprByteCodeObject UnlinkedBCO {..} + = multiLineEntry (text "ordinary bytecode object:") $ + vcat [ + {- + unlinkedBCOName :: !Name, + unlinkedBCOArity :: {-# UNPACK #-} !Int, + unlinkedBCOInstrs :: !(BCOByteArray Word16), -- insns + unlinkedBCOBitmap :: !(BCOByteArray Word), -- bitmap + unlinkedBCOLits :: !(FlatBag BCONPtr), -- non-ptrs + unlinkedBCOPtrs :: !(FlatBag BCOPtr) -- ptrs + -} + ] +pprByteCodeObject UnlinkedStaticCon {..} + = multiLineEntry (text "static constructor object:") $ + vcat [ + {- + unlinkedStaticConName :: !Name, + -- ^ The name to which this static constructor is bound, not to be + -- confused with the name of the static constructor itself + -- ('unlinkedStaticConDataConName') + unlinkedStaticConDataConName :: !Name, + unlinkedStaticConLits :: !(FlatBag BCONPtr), + -- ^ non-ptrs full words, where sub-word literals have already been + -- packed into full words as needed + unlinkedStaticConPtrs :: !(FlatBag BCOPtr), -- ptrs + unlinkedStaticConIsUnlifted :: !Bool + -} + ] -- | […] pprDataConstructorInfoTables :: [(Name, ConInfoTable)] -> SDoc -pprDataConstructorInfoTables = pprDataConstructorInfoTables +pprDataConstructorInfoTables + = multiLineEntry (text "data constructor info tables:") . + pprDataConstructorInfoTables -- | […] pprTopLevelStrings :: [(Name, ByteString)] -> SDoc -pprTopLevelStrings = pprTopLevelStrings +pprTopLevelStrings = multiLineEntry (text "top-level strings:") . + pprTopLevelStrings + +-- | […] +pprBreakPoints :: Maybe InternalModBreaks -> SDoc +pprBreakPoints = multiLineEntry (text "break points:") . + maybe (text "<none>") pprBreakPointsData -- | […] -pprInternalBreakPoints :: InternalModBreaks -> SDoc -pprInternalBreakPoints = pprInternalBreakPoints +pprBreakPointsData :: InternalModBreaks -> SDoc +pprBreakPointsData = pprBreakPointsData +{-FIMXE: + Instead of outputting module perhaps assert that it is identical to the + present one (if this indeed how it should be). +-} -- | […] pprStaticPointerTableEntries :: [SptEntry] -> SDoc -pprStaticPointerTableEntries = pprStaticPointerTableEntries +pprStaticPointerTableEntries + = multiLineEntry (text "static-pointer table entries:") . + pprStaticPointerTableEntries + +-- | […] +pprHPCInfo :: Strict.Maybe ByteCodeHpcInfo -> SDoc +pprHPCInfo = multiLineEntry (text "HPC information:") . + Strict.maybe (text "<none>") pprHPCInfoData -- | […] -pprHPCInfo :: ByteCodeHpcInfo -> SDoc -pprHPCInfo = pprHPCInfo +pprHPCInfoData :: ByteCodeHpcInfo -> SDoc +pprHPCInfoData = pprHPCInfoData -- | […] pprObjectFileContents :: [ByteString] -> SDoc -pprObjectFileContents = vcatOrNone . zipWith pprEntry [0 ..] where +pprObjectFileContents + = multiLineEntry (text "contents of object files:") . + vcatOrNone . + zipWith pprEntry [0 ..] + where pprEntry :: Int -> ByteString -> SDoc pprEntry ix content = multiLineEntry (text "file" <+> ppr ix) View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/4f217a585c7e4c142b7f8434a9e2ca… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/4f217a585c7e4c142b7f8434a9e2ca… You're receiving this email because of your account on gitlab.haskell.org.

1 0