August 2025 - ghc-commits

[Git][ghc/ghc][wip/T21730-import] Don't warn unused-imports with used generated imports (ref #21730)
by Brandon Chinn (＠brandonchinn178) 19 Aug '25

19 Aug '25

Brandon Chinn pushed to branch wip/T21730-import at Glasgow Haskell Compiler / GHC Commits: 9d8bdab4 by Brandon Chinn at 2025-08-19T08:38:55-07:00 Don't warn unused-imports with used generated imports (ref #21730) - - - - - 61 changed files: - compiler/GHC/Driver/Backpack.hs - compiler/GHC/Hs/ImpExp.hs - compiler/GHC/Iface/Ext/Ast.hs - compiler/GHC/Parser/Header.hs - compiler/GHC/Rename/Names.hs - compiler/GHC/Tc/Module.hs - compiler/GHC/Types/Name/Reader.hs - ghc/GHCi/UI.hs - testsuite/tests/gadt/T12468.stderr - testsuite/tests/ghc-e/should_fail/T24172.stderr - testsuite/tests/ghci/scripts/T8353.stderr - testsuite/tests/module/mod150.stderr - testsuite/tests/module/mod151.stderr - testsuite/tests/module/mod152.stderr - testsuite/tests/module/mod153.stderr - testsuite/tests/overloadedrecflds/should_fail/T18999_NoDisambiguateRecordFields.stderr - testsuite/tests/parser/should_compile/DumpRenamedAst.stderr - testsuite/tests/parser/should_compile/T14189.stderr - testsuite/tests/parser/should_compile/T19082.stderr - testsuite/tests/perf/compiler/hard_hole_fits.stderr - testsuite/tests/plugins/Makefile - + testsuite/tests/plugins/T21730-plugin/Makefile - + testsuite/tests/plugins/T21730-plugin/Setup.hs - + testsuite/tests/plugins/T21730-plugin/T21730-plugin.cabal - + testsuite/tests/plugins/T21730-plugin/T21730_Plugin.hs - + testsuite/tests/plugins/T21730.hs - testsuite/tests/plugins/all.T - testsuite/tests/quotes/LiftErrMsg.stderr - testsuite/tests/quotes/LiftErrMsgDefer.stderr - testsuite/tests/quotes/LiftErrMsgTyped.stderr - testsuite/tests/rename/should_compile/T22513d.stderr - testsuite/tests/rename/should_compile/T22513e.stderr - testsuite/tests/rename/should_compile/T22513f.stderr - testsuite/tests/rename/should_compile/T22513g.stderr - testsuite/tests/rename/should_compile/T22513h.stderr - testsuite/tests/rename/should_compile/T22513i.stderr - testsuite/tests/rename/should_compile/rn039.ghc.stderr - testsuite/tests/rename/should_fail/T15487.stderr - testsuite/tests/rename/should_fail/T18740a.stderr - testsuite/tests/rename/should_fail/rnfail044.stderr - testsuite/tests/safeHaskell/flags/SafeFlags17.stderr - testsuite/tests/th/T10267.stderr - testsuite/tests/th/T14627.stderr - testsuite/tests/th/T15321.stderr - testsuite/tests/typecheck/should_compile/T13050.stderr - testsuite/tests/typecheck/should_compile/T14273.stderr - testsuite/tests/typecheck/should_compile/T14590.stderr - testsuite/tests/typecheck/should_compile/T25180.stderr - testsuite/tests/typecheck/should_compile/T9497a.stderr - testsuite/tests/typecheck/should_compile/refinement_hole_fits.stderr - testsuite/tests/typecheck/should_compile/subsumption_sort_hole_fits.stderr - testsuite/tests/typecheck/should_fail/T14884.stderr - testsuite/tests/typecheck/should_fail/T21130.stderr - testsuite/tests/typecheck/should_fail/T23739b.stderr - testsuite/tests/typecheck/should_fail/T23739c.stderr - testsuite/tests/typecheck/should_fail/T9497d.stderr - testsuite/tests/typecheck/should_fail/tcfail037.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 - testsuite/tests/vdq-rta/should_fail/T23738_fail_pun.stderr The diff was not included because it is too large. View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/9d8bdab48b43dfb312bad9fa835c13a… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/9d8bdab48b43dfb312bad9fa835c13a… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc] Pushed new branch revert-0a583689
by Ben Gamari (＠bgamari) 19 Aug '25

19 Aug '25

Ben Gamari pushed new branch revert-0a583689 at Glasgow Haskell Compiler / GHC -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/tree/revert-0a583689 You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/T26312] 2 commits: rts/IPE: Fix compilation when zstd is enabled
by Ben Gamari (＠bgamari) 19 Aug '25

19 Aug '25

Ben Gamari pushed to branch wip/T26312 at Glasgow Haskell Compiler / GHC Commits: 9767adb7 by Ben Gamari at 2025-08-19T11:36:29-04:00 rts/IPE: Fix compilation when zstd is enabled This was broken by the refactoring undertaken in c80dd91c0bf6ac034f0c592f16c548b9408a8481. Closes #26312. - - - - - 06714855 by Ben Gamari at 2025-08-19T11:37:01-04:00 hadrian: Pass lib & include directories to ghc `Setup configure` - - - - - 2 changed files: - hadrian/src/Settings/Packages.hs - rts/IPE.c Changes: ===================================== hadrian/src/Settings/Packages.hs ===================================== @@ -36,6 +36,8 @@ packageArgs = do cursesLibraryDir <- getSetting CursesLibDir ffiIncludeDir <- getSetting FfiIncludeDir ffiLibraryDir <- getSetting FfiLibDir + libzstdIncludeDir <- getSetting LibZstdIncludeDir + libzstdLibraryDir <- getSetting LibZstdLibDir stageVersion <- readVersion <$> (expr $ ghcVersionStage stage) mconcat @@ -74,7 +76,9 @@ packageArgs = do , builder (Cabal Setup) ? mconcat [ arg "--disable-library-for-ghci" , anyTargetOs [OSOpenBSD] ? arg "--ld-options=-E" - , compilerStageOption ghcProfiled ? arg "--ghc-pkg-option=--force" ] + , compilerStageOption ghcProfiled ? arg "--ghc-pkg-option=--force" + , cabalExtraDirs libzstdIncludeDir libzstdLibraryDir + ] , builder (Cabal Flags) ? mconcat -- For the ghc library, internal-interpreter only makes ===================================== rts/IPE.c ===================================== @@ -314,7 +314,7 @@ void decompressIPEBufferListNodeIfCompressed(IpeBufferListNode *node) { "decompression library (zstd) is not available."); #else size_t compressed_sz = ZSTD_findFrameCompressedSize( - node->string_table, + node->string_table_block->string_table, node->string_table_size ); char *decompressed_strings = stgMallocBytes( @@ -324,10 +324,10 @@ void decompressIPEBufferListNodeIfCompressed(IpeBufferListNode *node) { ZSTD_decompress( decompressed_strings, node->string_table_size, - node->string_table, + node->string_table_block->string_table, compressed_sz ); - node->string_table = (const char *) decompressed_strings; + node->string_table_block->string_table = (const char *) decompressed_strings; // Decompress the IPE data compressed_sz = ZSTD_findFrameCompressedSize( View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/d894fa3e5b62b3df19442b954c611d… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/d894fa3e5b62b3df19442b954c611d… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc][wip/T26312] 2 commits: rts/IPE: Fix compilation when zstd is enabled
by Ben Gamari (＠bgamari) 19 Aug '25

19 Aug '25

Ben Gamari pushed to branch wip/T26312 at Glasgow Haskell Compiler / GHC Commits: 1b98c26b by Ben Gamari at 2025-08-19T11:26:11-04:00 rts/IPE: Fix compilation when zstd is enabled - - - - - d894fa3e by Ben Gamari at 2025-08-19T11:26:11-04:00 hadrian: Pass lib & include directories to ghc `Setup configure` - - - - - 2 changed files: - hadrian/src/Settings/Packages.hs - rts/IPE.c Changes: ===================================== hadrian/src/Settings/Packages.hs ===================================== @@ -36,6 +36,8 @@ packageArgs = do cursesLibraryDir <- getSetting CursesLibDir ffiIncludeDir <- getSetting FfiIncludeDir ffiLibraryDir <- getSetting FfiLibDir + libzstdIncludeDir <- getSetting LibZstdIncludeDir + libzstdLibraryDir <- getSetting LibZstdLibDir stageVersion <- readVersion <$> (expr $ ghcVersionStage stage) mconcat @@ -74,7 +76,9 @@ packageArgs = do , builder (Cabal Setup) ? mconcat [ arg "--disable-library-for-ghci" , anyTargetOs [OSOpenBSD] ? arg "--ld-options=-E" - , compilerStageOption ghcProfiled ? arg "--ghc-pkg-option=--force" ] + , compilerStageOption ghcProfiled ? arg "--ghc-pkg-option=--force" + , cabalExtraDirs libzstdIncludeDir libzstdLibraryDir + ] , builder (Cabal Flags) ? mconcat -- For the ghc library, internal-interpreter only makes ===================================== rts/IPE.c ===================================== @@ -314,7 +314,7 @@ void decompressIPEBufferListNodeIfCompressed(IpeBufferListNode *node) { "decompression library (zstd) is not available."); #else size_t compressed_sz = ZSTD_findFrameCompressedSize( - node->string_table, + node->string_table_block->string_table, node->string_table_size ); char *decompressed_strings = stgMallocBytes( @@ -324,10 +324,10 @@ void decompressIPEBufferListNodeIfCompressed(IpeBufferListNode *node) { ZSTD_decompress( decompressed_strings, node->string_table_size, - node->string_table, + node->string_table_block->string_table, compressed_sz ); - node->string_table = (const char *) decompressed_strings; + node->string_table_block->string_table = (const char *) decompressed_strings; // Decompress the IPE data compressed_sz = ZSTD_findFrameCompressedSize( View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/67d38774f1f5779fb0051ba5f6af58… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/67d38774f1f5779fb0051ba5f6af58… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc] Pushed new branch wip/T26312
by Ben Gamari (＠bgamari) 19 Aug '25

19 Aug '25

Ben Gamari pushed new branch wip/T26312 at Glasgow Haskell Compiler / GHC -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/tree/wip/T26312 You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc] Pushed new branch wip/T26316
by Ben Gamari (＠bgamari) 19 Aug '25

19 Aug '25

Ben Gamari pushed new branch wip/T26316 at Glasgow Haskell Compiler / GHC -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/tree/wip/T26316 You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc] Pushed new branch wip/romes/26325
by Rodrigo Mesquita (＠alt-romes) 19 Aug '25

19 Aug '25

Rodrigo Mesquita pushed new branch wip/romes/26325 at Glasgow Haskell Compiler / GHC -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/tree/wip/romes/26325 You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc] Pushed new branch wip/backports-9.14
by Ben Gamari (＠bgamari) 19 Aug '25

19 Aug '25

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

1 0

[Git][ghc/ghc][wip/T26053] 2 commits: nonmoving: Use get_itbl instead of explicit loads
by Ben Gamari (＠bgamari) 19 Aug '25

19 Aug '25

Ben Gamari pushed to branch wip/T26053 at Glasgow Haskell Compiler / GHC Commits: 7e7213bd by Ben Gamari at 2025-08-19T10:21:24-04:00 nonmoving: Use get_itbl instead of explicit loads This is cleaner and also fixes unnecessary (and unsound) use of `volatile`. - - - - - a84b4bae by Ben Gamari at 2025-08-19T10:21:24-04:00 rts/Scav: Handle WHITEHOLEs in scavenge_one `scavenge_one`, used to scavenge mutable list entries, may encounter `WHITEHOLE`s when the non-moving GC is in use via two paths: 1. when an MVAR is being marked concurrently 2. when the object belongs to a chain of selectors being short-cutted. Fixes #26204. - - - - - 2 changed files: - rts/sm/NonMovingMark.c - rts/sm/Scav.c Changes: ===================================== rts/sm/NonMovingMark.c ===================================== @@ -619,7 +619,7 @@ inline void updateRemembSetPushThunk(Capability *cap, StgThunk *thunk) { const StgInfoTable *info; do { - info = *(StgInfoTable* volatile*) &thunk->header.info; + info = (StgInfoTable*) RELAXED_LOAD(&thunk->header.info); } while (info == &stg_WHITEHOLE_info); const StgThunkInfoTable *thunk_info = THUNK_INFO_PTR_TO_STRUCT(info); @@ -1343,7 +1343,7 @@ mark_closure (MarkQueue *queue, const StgClosure *p0, StgClosure **origin) goto done; case WHITEHOLE: - while (*(StgInfoTable* volatile*) &p->header.info == &stg_WHITEHOLE_info) + while (RELAXED_LOAD(&p->header.info) == &stg_WHITEHOLE_info) #if defined(PARALLEL_GC) busy_wait_nop() #endif @@ -1714,7 +1714,7 @@ mark_closure (MarkQueue *queue, const StgClosure *p0, StgClosure **origin) break; case WHITEHOLE: - while (*(StgInfoTable* volatile*) &p->header.info == &stg_WHITEHOLE_info); + while ((StgInfoTable *) RELAXED_LOAD(&p->header.info) == &stg_WHITEHOLE_info); goto try_again; case COMPACT_NFDATA: ===================================== rts/sm/Scav.c ===================================== @@ -1276,6 +1276,7 @@ scavenge_one(StgPtr p) bool no_luck; bool saved_eager_promotion; +try_again: saved_eager_promotion = gct->eager_promotion; ASSERT(LOOKS_LIKE_CLOSURE_PTR(p)); @@ -1617,6 +1618,13 @@ scavenge_one(StgPtr p) scavenge_continuation((StgContinuation *)p); break; + case WHITEHOLE: + // This may happen when the nonmoving GC is in use via two paths: + // 1. when an MVAR is being marked concurrently + // 2. when the object belongs to a chain of selectors being short-cutted. + while (get_itbl((StgClosure*) p) == &stg_WHITEHOLE_info); + goto try_again; + default: barf("scavenge_one: strange object %d", (int)(info->type)); } View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/d365d9311666f0a0762b38e5640f1e… -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/d365d9311666f0a0762b38e5640f1e… You're receiving this email because of your account on gitlab.haskell.org.

1 0

[Git][ghc/ghc] Pushed new branch wip/bump-min-ghc
by Ben Gamari (＠bgamari) 19 Aug '25

19 Aug '25

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

1 0