Closed
Milestone
7.2.1
Unstarted Issues (open and unassigned)
0
Ongoing Issues (open and assigned)
0
Completed Issues (closed)
237
- panic after pressing Ctrl+C in ghci
- Regression in newName
- ghc: panic! mkUsageInfo: internal name? constr_a2Py{v}
- GHC 7.2.1-rc1 panics on `cgLookupPanic`
- Cabal bug building haddock, alex, happy, etc
- 2047 core lint error when profiling
- signals004(profasm) core lint error
- Annotations aren't working
- Windows build broken
- No read permission for installed global package.cache
- Problems with new cyclic dependency error message
- Core lint failure in stage 3 build
- panic: getPredTyDescription EqPred [7.0 regression]
- Core lint error in T3001-2 [7.0 regression]
- Installing ghc ruins PATH on Windows
- Non-understandable comment in compiler/main/GHC.hs
- unregisterised build broken on arm-linux platform
- RTS build failure with gcc-4.6.1
- Switch to upstream binary repo
- stage3 build failing with core-lint error
- installing xcode 4 breaks ghc linking
- ghc-asm has a poor shebang
- ghc-7 produces 10 times bigger binaries on macs than ghc-6
- epollCreate: unsupported operation (Function not implemented)
- 7.0.2 ignores a context which 7.0.1 picks up
- Mac installers are missing ghc & cabal guides
- Warning about unqualified implicit imports
- building ghc7.0.1.20110217 under x86 solaris fails
- Merge plugins into HEAD
- SRTs bigger than they should be?
- OS X, XCode 4 and gcc
- panic in idInfoToAmode with -O
- "internal error: heapCencus, unknown object: 0" with retainer profiling
- wrong dylib name using GHC package
- GHC crash (panic) with -O2 and nonexistent .hi
- ghc.exe: panic! (the 'impossible' happened)
- annrun01 needs to be updated
- framework failure: duplicate T1595
- Bad compiler warning
- Data.Typeable not backwards compatible
- Heap used by compacting collector larger than heap used without.
- Calling fail on a UTF-8 encoded string (in file) causes garbage to be printed
- SEGFAULT in FFI to C++ library
- ghc --make and ghci misses dependencies with explicit braces
- Wrong format specifiers break debug build
- Large space usage when deriving Generic
- Typos in GHC documentation
- Typos in future POSIX process group API implementation
- unicode regression
- process library does not build with GHC 7.0.x
- Add primops for copying/cloning byte arrays
- Stack overflow in criterion
- createDirectory does not retry on EINTR (interrupted)
- dry run option -n broken
- Unexpected Failure due to change of defaultDynFlags' type
- RULES "minusFloat x x" and "timesFloat 0.0 x" break IEEE-754 compatibility
- Segfault in reportCCSProfiling
- Implement forward substitution of constants in the Cmm mini-inliner
- Poor performance of division; unnecessary branching
- ghci :set +s yields absurd negative times
- ghc 7.1 says "Use -fvia-C instead" but -fvia-C is gone
- SpecConstr messages should be reworded, and not be shown at standard verbosity level
- inferred type of an implicit parameter rejected (associated type)
- -with-rtsopts, ghc_rts_opts not honoured
- Build system is not generating dependencies for the RTS
- Stack space overflow when using -N2 and not with -N1
- print026 sporadically fails
- validate should bail out early if testsuite not available
- Recompilation bug: recomp008 failed
- No instance Show [: Int :] in dph-diophantine-opt
- testsuite failures: hpc_markup_multi_*
- OS X: hClose003
- dph tests failing
- Implement FFI spec behaviour for *CString family
- iteratee: epollControl: permission denied (Operation not permitted)
- getProcessExitCode doesn't retry on EINTR
- typo in DEBUG RTS options
- Internal error with runhaskell and System.Plugins.dynload
- Produces invalid assembly when using `peek' and -fvia-c with new binutils on amd64
- GHC Panic on Text.Highlighting.Kate.Syntax.Ocaml
- Wrong SrcSpan on AbsBinds
- ghc fails to build if HOME is not seth
- "panic! (the 'impossible' happened)" compiling code using arrows and elimReader (test case included)
- Compiling bytestring-mmap with -fvia-C yields broken .s file, two .size directives
- STM, newArray, and a stack overflow
- internal error with rebindable if-then-else
- ghc bug: urk! lookup local fingerprint
- Performance of Data.Graph.{preorderF, postorderF}
- Slow type checking of type-level computation heavy code.
- Remove ability to use ghc-binary
- +RTS -G1 is broken
- Users Guide formatting bug
- "deriving" seems to ignore class context for a type family
- test 4038 (ghci) fails on OS X 64 bit.
- build fails on powerpc: error: 'ObjectCode' has no member named 'misalignment'
- performance regressions
- Compiling pandoc with llvm backend fails with panic
- build error on mingw if pthreads is installed
- darcs: internal error: stg_ap_v_ret (GHC version 7.0.1 for x86_64_apple_darwin)
- OS X: ld: warning: -read_only_relocs cannot be used with x86_64
- inconsistent class requirements with TypeFamilies and FlexibleContexts
- Continuation passing style loop doesn't compile into a loop
- time002 and time004 (ghci) test failures on OS X 64 bit
- internal error: stg_ap_v_ret on forkProcess + executeFile
- 60% performance regression in continuation-heavy code between 6.12 and 7
- Dead code fed to CorePrep because RULEs keep it alive spuriously
- threadDelay behavior 64-bit mac os x
- Regenerated cbits/WCsubst.c based on Unicode 6.0.0
- reify does not work on data structures defined in files without Language pragma of Template Haskell and without -fth or -XTemplateHaskell command line flags
- Inliner looping when specialising across modules (with GADTs and other extensions)
- panic building darcs
- Document behaviour of throwTo to your own ThreadId
- registerPackage fails with multiple command line --package-conf= flags
- ghc misdiagnoses a compile time error concerning parameterized types
- ghci command line option -l should accept either -llibrary or -l library (POSIX requirement)
- Debug.Trace.trace mangles Unicode strings
- NoImplicitPrelude does not handle rec / mfix / ArrowLoop properly
- LLVM mangler takes too long at runtime
- Inconsistent import of instances in GHCi
- ghci fails to load fat binary archives on OS X
- DLLs not documented consistently
- Versions of bracket without masking acquire action
- ghc-stage2: Monadic.o: bus error / segmentation fault / internal error
- Make SampleVar an instance of Typeable
- Deriving Data for existentially quantified types
- Do not link in unused code
- Over eager non-exhaustive pattern match warnings in lambda expressions.
- No links will be generated to these packages: base-4.3.0.0
- dataToQa uses only unqualified names when converting values to their TH representation.
- The new flag -fwarn-missing-import-lists does not recognize the (..) form of import.
- Unable to satisfy "Defined but not used: data constructor" warning
- Per-class incoherence, and solve incoherent instances last
- Allow type variables to be instantiated as a typeclass
- Exporting the same entity twice should not be a warning
- unreliable ghc on sparc
- 3 ways to write a function (unexpected performance difference and regression)
- GHC --make switch does not set file timestamp correctly
- RTS options broken for dynamic libraries
- Missing atexit symbol
- NCG miscompiles Double -> Float -> Double
- T3016 failed with timeout (hpc and optasm)
- barton-mangler-bug failed with timeout (multiple ways)
- Remove 'rename' and friends from TH.Lib
- dph-words-opt fails to compile
- Asynchronous exceptions in runInUnboundThread
- GHCi says: ghc: internal error: evacuate: strange closure type 63587
- forkIO threads do not properly save/restore the floating point environment
- Unexpected failures, too much allocation
- Lexer does not handle unicode numeric subscripts
- weird discrepancies between TFs and FDs in GHC7
- fromRational broken for Ratio a
- Crash while building HEAD on OS X
- Deferred equalities and forall types
- Impossible when deriving empty data declaration
- -O0 runs in constant space, -O1 and -O2 don't
- Out of memory when compiling Statistics.Quantile
- GHC's runtime never terminates unused worker threads
- Poor error message when openFile fails to open named pipe
- getCPUTime on x86_64 Mac OS X 10.6
- sequence is not tail recursive, doesn't work with large inputs in strict monads
- LLVM: space_leak_001 fails when GHC built with LLVM backend
- -dcore-lint and -fstrict-dicts break anything that uses a type class
- broken links in Cabal docs
- GHC does not accept --srcdir
- Bad choice of loop breaker?
- System.Process doesn't encode its arguments.
- '+RTS -S' reports negative allocation.
- Require XOverlappingInstances for the most specific instance only
- GHC leaks memory when compiling many files
- Problem with wxHaskell
- SplitObjs fails on sparc with GNU ld
- Missing -lz option in testsuite
- type checker fails to infer an implicit parameter constraint in the presence of existential types.
- inlining happens on foldl1 and does not happen on direct application of combinator
- GHC API no longer exports location information for error/warning messages
- Distinguish type parameters from indices
- Code compiled WITHOUT profiling many times slower than compiled WITH profiling on
- Unreachable code in optimised CMM code
- Inliner behaviour with instances is confusing
- darcs-all cannot use SSH repos with MSYS
- :step hangs when -fbreak-on-exception is set
- the recompilation checker should take CPP flags into account
- getArgs should return Unicode on Unix
- getArgs should return Unicode on Windows
- System.IO and System.Directory functions not Unicode-aware under Unix
- define gcd 0 0 = 0
- Compiler panic on incorrect code (TcTyFuns.flattenType: synonym family in a rank-n type)
- :history throws "Irrefutable pattern failed" exception
- ASSERT failed: file typecheck/TcMType.lhs line 526 t_a32G{tv}
- Error messages have the wrong qualified names
- Example where ghc 6.10.1 fails to optimize recursive instance function calls
- Excessive heap usage
- Equality constraint in superclass not supported
- Stack overflow in stream-fusion library with GHC 6.10 RC
- Data.Time.Clock parseTime's %y parses "08" as 1908
- Unknown symbol `_environ' on MacOS X
- Bad warning when exporting data families
- properFraction implemented with modf primitive?
- floor, ceiling, round :: Double -> Int are awesomely slow
- Numeric literal printed wrong in error message
- Deep stacks make execution time go through the roof
- Typeclass membership doesn't bring coercion superclass requirements into scope
- Ambiguous types and rejected type signatures
- Retain export-list order in ModIface, use it in :browse
- -Werror aborts too early
- seq can make code slower
- duplicate "not in scope" error when giving multiple vars type-signatures at once
- CString marshalling functions do not perform the specified conversion
- Add hPutNonBlocking to Data.ByteString
- GHC uses _stub.c files regardless of whether any 'foreign import' decls remain in a .hs file
- Recompilation check fails for TH
- Improve the API for TChan, TMVar, and TVar
- Deprecate Control.Exception.blocked in favour of getMaskingState
- Make the Timeout exception a newtype instead of a datatype
- Some patches for GHC
- add no spaces allowed in new qualified operators to user guide
- Build the ghc-bin package in the standard way
- Easily make Typeable keys pure, so that Typeable can be handled efficiently across communications
- -Werror should wait until type-checking is done
- New POSIX process group API
- GHC should support -x objective-cpp, as well
- Output ELF .size directives
- GHC and GHCi use different syntax for dynamic libraries
- GHC should support -x objective-c
- Add primops for copying/cloning an array in the new codegen
- Make Chan an instance of Eq
- Warn about unnecessary fromIntegral and other conversions
- Way to determine if code is running in thread fixed on capability
- Better support for resolving infix expressions in template haskell
- Quasiquoting without bloating
- Default to all cores for executables compiled with new flag -multicore
- allow implicit parameter bindings in patterns
- Merge _stub.o files back in to the .o file
- Support reading .a files in GHCi to reclaim some disk space
- implement the Cabal ${pkgroot} spec extension
- having to call hs_add_root(__stginit_Foo) is a bit of a pain
- Better stack management please
