Marge Bot pushed to branch master at Glasgow Haskell Compiler / GHC Commits: 7b616b9f by Cheng Shao at 2026-01-16T06:45:00-05:00 compiler: fix regression when compiling foreign stubs in the rts unit This patch fixes a regression when compiling foreign stubs in the rts unit introduced in 05e25647f72bc102061af3f20478aa72bff6ff6e. A simple revert would fix it, but it's better to implement a proper fix with comment for better understanding of the underlying problem, see the added comment for explanation. Co-authored-by: Codex <codex@openai.com> - - - - - 1 changed file: - compiler/GHC/Driver/CodeOutput.hs Changes: ===================================== compiler/GHC/Driver/CodeOutput.hs ===================================== @@ -296,10 +296,15 @@ outputForeignStubs logger tmpfs dflags unit_state mod location stubs mk_include i = "#include \"" ++ ST.unpack i ++ "\"\n" in case mrts_pkg of Just rts_pkg -> concatMap mk_include (unitIncludes rts_pkg) - -- This case only happens when compiling foreign stub for the rts - -- library itself. The only time we do this at the moment is for - -- IPE information for the RTS info tables - Nothing -> "" + -- The Nothing case only happens when compiling + -- foreign stubs for the rts library itself (e.g. + -- building with +ipe), and the rts unit is not + -- registered yet. + -- + -- The generated stubs may still use RTS API, so + -- we must ensure that Rts.h is included, + -- otherwise we may run into regressions (#26779). + Nothing -> "#include \"Rts.h\"\n" -- wrapper code mentions the ffi_arg type, which comes from ffi.h ffi_includes View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/7b616b9f0236f4160f28923784b448f3... -- View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/7b616b9f0236f4160f28923784b448f3... You're receiving this email because of your account on gitlab.haskell.org.