June 2017 - ghc-tickets

[GHC] #9281: Rewrite `integer-gmp` to use only non-allocating GMP functions
by GHC 20 Sep '17

20 Sep '17

#9281: Rewrite `integer-gmp` to use only non-allocating GMP functions -------------------------------------+------------------------------------- Reporter: hvr | Owner: hvr Type: task | Status: new Priority: normal | Milestone: 7.10.1 Component: libraries (other) | Version: Keywords: integer-gmp | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | -------------------------------------+------------------------------------- After some promising results with a proof-of-concept implementation, I'm optimistic we can rewrite `integer-gmp` to use only non-allocating GMP lib functions without suffering from serious regressions. If successful, this would - allow to avoid the custom GMP allocator hack, and thus - avoid issues when linking against other C libraries using GMP, - simplify code, as we would perform all heap allocations in Haskell code (and never inside Cmm/C code as its done now), - and finally maybe even remove a few more superfluous temporary heap allocations. This rewrite can be done incrementally. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9281> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 32

[GHC] #13481: T12622 fails in ghci way
by GHC 20 Sep '17

20 Sep '17

#13481: T12622 fails in ghci way -------------------------------------+------------------------------------- Reporter: bgamari | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: 8.2.1 Component: Compiler | Version: 8.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Compile-time Unknown/Multiple | crash or panic Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- `T12622`, which is intended to test StaticPointers, fails in the GHCi way with a core lint warning, {{{ *** Core Lint errors : in result of Float out(FOS {Lam = Just 0, Consts = True, OverSatApps = False}) *** <no location info>: warning: In the expression: >>= @ IO $fMonadIO @ (Bool -> Bool) @ () (break<10>(s_a1DF) lvl_s3bD) lvl_s3bE s_a1DF :: StaticPtr (Bool -> Bool) [LclId, Unf=Unf{Src=<vanilla>, TopLvl=False, Value=False, ConLike=False, WorkFree=False, Expandable=False, Guidance=IF_ARGS [] 70 0}] is out of scope *** Offending Program *** g :: Bool [LclIdX] g = break<15>() True $trModule_s3b7 :: Addr# [LclId] $trModule_s3b7 = "main"# $trModule_s3b6 :: TrName [LclId] $trModule_s3b6 = TrNameS $trModule_s3b7 $trModule_s3b9 :: Addr# [LclId] $trModule_s3b9 = "Main"# $trModule_s3b8 :: TrName [LclId] $trModule_s3b8 = TrNameS $trModule_s3b9 $trModule :: Module [LclIdX] $trModule = Module $trModule_s3b6 $trModule_s3b8 lvl_s3bj :: Addr# [LclId] lvl_s3bj = "error"# lvl_s3bk :: [Char] [LclId] lvl_s3bk = unpackCString# lvl_s3bj lvl_s3bl :: Addr# [LclId] lvl_s3bl = "main"# lvl_s3bm :: [Char] [LclId] lvl_s3bm = unpackCString# lvl_s3bl lvl_s3bn :: Addr# [LclId] lvl_s3bn = "Main"# lvl_s3bo :: [Char] [LclId] lvl_s3bo = unpackCString# lvl_s3bn lvl_s3bp :: Addr# [LclId] lvl_s3bp = "T12622.hs"# lvl_s3bq :: [Char] [LclId] lvl_s3bq = unpackCString# lvl_s3bp lvl_s3br :: Int [LclId] lvl_s3br = I# 21# lvl_s3bs :: Int [LclId] lvl_s3bs = I# 14# lvl_s3bt :: Int [LclId] lvl_s3bt = I# 21# lvl_s3bu :: Int [LclId] lvl_s3bu = I# 64# lvl_s3bv :: SrcLoc [LclId] lvl_s3bv = SrcLoc lvl_s3bm lvl_s3bo lvl_s3bq lvl_s3br lvl_s3bs lvl_s3bt lvl_s3bu lvl_s3bw :: ([Char], SrcLoc) [LclId] lvl_s3bw = (lvl_s3bk, lvl_s3bv) $dIP_s3bc :: CallStack [LclId] $dIP_s3bc = pushCallStack lvl_s3bw emptyCallStack lvl_s3bx :: Addr# [LclId] lvl_s3bx = "couldn't find "# lvl_s3by :: [Char] [LclId] lvl_s3by = unpackCString# lvl_s3bx lookupKey :: forall a. StaticPtr a -> IO a [LclIdX, Arity=1] lookupKey = \ (@ a_a23X) (p_X1DX :: StaticPtr a_a23X) -> break<8>(p_X1DX) >>= @ IO $fMonadIO @ (Maybe (StaticPtr a_a23X)) @ a_a23X (break<1>(p_X1DX) unsafeLookupStaticPtr @ a_a23X (break<0>(p_X1DX) staticKey @ a_a23X p_X1DX)) (\ (ds_d3as :: Maybe (StaticPtr a_a23X)) -> case ds_d3as of { Nothing -> break<7>(p_X1DX) error @ 'LiftedRep @ (IO a_a23X) ($dIP_s3bc `cast` (Sym (N:IP[0] <"callStack">_N <CallStack>_N) :: (CallStack :: *) ~R# ((?callStack::CallStack) :: Constraint))) (break<6>(p_X1DX) ++ @ Char lvl_s3by (break<5>(p_X1DX) show @ StaticPtrInfo $fShowStaticPtrInfo (break<4>(p_X1DX) staticPtrInfo @ a_a23X p_X1DX))); Just p_a1DI -> break<3>(p_a1DI) return @ IO $fMonadIO @ a_a23X (break<2>(p_a1DI) deRefStaticPtr @ a_a23X p_a1DI) }) $dTypeable_s3bA :: TypeRep Bool [LclId] $dTypeable_s3bA = mkTrCon @ * @ Bool $tcBool ([] @ SomeTypeRep) s_s3bC :: StaticPtr (Bool -> Bool) [LclId] s_s3bC = case break<9>() sg @ Bool ($dTypeable_s3bA `cast` (Sym N:Typeable[0] <*>_N <Bool>_N :: (TypeRep Bool :: *) ~R# (Typeable Bool :: Constraint))) of { T s_a38b -> s_a38b } lvl_s3bD :: IO (Bool -> Bool) [LclId] lvl_s3bD = lookupKey @ (Bool -> Bool) s_s3bC lvl_s3bE :: (Bool -> Bool) -> IO () [LclId] lvl_s3bE = \ (f_a1DG :: Bool -> Bool) -> break<12>(f_a1DG) print @ Bool $fShowBool (break<11>(f_a1DG) f_a1DG True) main :: IO () [LclIdX] main = break<14>() break<13>(s_s3bC) >>= @ IO $fMonadIO @ (Bool -> Bool) @ () (break<10>(s_a1DF) lvl_s3bD) lvl_s3bE main :: IO () [LclIdX] main = runMainIO @ () main *** End of Offense *** <no location info>: error: Compilation had errors *** Exception: ExitFailure 1 ===== program output begins here ===== program output begins here T12622:6:30: error: Not in scope: ‘Main.main’ No module named ‘Main’ is imported. }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13481> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 12

[GHC] #13525: hWaitForInput with timeout causes program to abort
by GHC 20 Sep '17

20 Sep '17

#13525: hWaitForInput with timeout causes program to abort --------------------------------------+---------------------------------- Reporter: bgamari | Owner: (none) Type: bug | Status: new Priority: high | Milestone: 8.2.1 Component: Compiler | Version: 8.0.2 Keywords: | Operating System: Linux Architecture: x86_64 (amd64) | Type of failure: Runtime crash Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: --------------------------------------+---------------------------------- This program, {{{#!hs import System.IO import System.Timeout main = hWaitForInput stdin (5 * 1000) }}} causes the program to abort (tested on Linux), {{{ $ ghc hi.hs [1 of 1] Compiling Main ( hi.hs, hi.o ) Linking hi ... $ ./hi fdReady: msecs != 0, this shouldn't happenAborted }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13525> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 10

[GHC] #9157: cmm common block not eliminated
by GHC 20 Sep '17

20 Sep '17

#9157: cmm common block not eliminated ------------------------------------+------------------------------------- Reporter: wojteknar | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.2 Keywords: | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: Other Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- For the toTuple# function (in the attached file) GHC generates a cmm switch statement, with all the alternatives, unsurprisingly, the same. Yet, cmm common block elimination does not kick in. In this particular example, the whole case statement could be annihilated, because all alternatives lead to the same code. BTW, this is how they match in France. {{{ #!ocaml type alt = A of int | B of int | C of int | D of int let g x = match x with A v | B v | C v | D v -> v }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9157> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 12

[GHC] #9775: "Failed to remove" errors during Windows build from hsc2hs
by GHC 18 Sep '17

18 Sep '17

#9775: "Failed to remove" errors during Windows build from hsc2hs -------------------------------------+------------------------------------- Reporter: gintas | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Build System | Version: 7.8.3 Keywords: | Operating System: Windows Architecture: Unknown/Multiple | Type of failure: Difficulty: Unknown | None/Unknown Blocked By: | Test Case: Related Tickets: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- The Windows build over here spews a bunch of errors during the build like the ones below. All of them seem to be related to hsc2hs. The errors do not the build for some reason, so the problem is not critical, but we should look into what's going on here. Note that the errors are indeterministic and the set of files affected varies between builds. Examples: {{{ ghctrace.2/3.log:Failed to remove file libraries/base/dist- install/build/System/CPUTime_hsc_make.exe; error= DeleteFile "libraries/base/dist-install/build/System/CPUTime_hsc_make.exe": permission denied (Access is denied.) ghctrace.2/4.log:Failed to remove file libraries/hpc/dist- boot/build/Trace/Hpc/Reflect_hsc_make.exe; error= DeleteFile "libraries/hpc/dist-boot/build/Trace/Hpc/Reflect_hsc_make.exe": permission denied (Access is denied.) ghctrace.2/4.log:Failed to remove file compiler/stage2/build/Fingerprint_hsc_make.exe; error= DeleteFile "compiler/stage2/build/Fingerprint_hsc_make.exe": permission denied (Access is denied.) ghctrace.3/1.log:Failed to remove file libraries/hpc/dist- boot/build/Trace/Hpc/Reflect_hsc_make.exe; error= DeleteFile "libraries/hpc/dist-boot/build/Trace/Hpc/Reflect_hsc_make.exe": permission denied (Access is denied.) ghctrace.3/4.log:Failed to remove file libraries/old-time/dist- install/build/System/Time_hsc_make.exe; error= DeleteFile "libraries/old- time/dist-install/build/System/Time_hsc_make.exe": permission denied (Access is denied.) ghctrace.3/5.log:Failed to remove file libraries/Win32/dist- install/build/Graphics/Win32/GDI/Pen_hsc_make.exe; error= DeleteFile "libraries/Win32/dist-install/build/Graphics/Win32/GDI/Pen_hsc_make.exe": permission denied (Access is denied.) }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9775> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 15

[GHC] #13767: GHCi trips -DS checks at rts/sm/Sanity.c, line 210
by GHC 18 Sep '17

18 Sep '17

#13767: GHCi trips -DS checks at rts/sm/Sanity.c, line 210 -------------------------------------+------------------------------------- Reporter: niteria | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Runtime | Version: 8.3 System | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Error: {{{ $ echo -e ":quit\n" | ./inplace/bin/ghc-stage2 --interactive +RTS -DS -RTS -ignore-dot-ghci GHCi, version 8.3.20170515: http://www.haskell.org/ghc/ :? for help Prelude> Leaving GHCi. ghc-stage2: internal error: ASSERTION FAILED: file rts/sm/Sanity.c, line 210 (GHC version 8.3.20170515 for x86_64_unknown_linux) Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug Aborted (core dumped) }}} Steps to reproduce: * Set `BuildFlavour = prof` in `mk/build.mk` * Add `GhcLibWays += p debug_p thr_debug_p` to `mk/build.mk`, I think you only need `thr_debug_p` * Build GHC * (cd ghc; make re2 GhcDebugged=YES), I think you can build it right away, but I'm not that familiar with the build system * `echo -e ":quit\n" | ./inplace/bin/ghc-stage2 --interactive +RTS -DS -RTS -ignore-dot-ghci` -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13767> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 9

[GHC] #10601: GHC should be distributed with debug symbols
by GHC 16 Sep '17

16 Sep '17

#10601: GHC should be distributed with debug symbols -------------------------------------+------------------------------------- Reporter: bitonic | Owner: Type: task | Status: new Priority: normal | Milestone: Component: Compiler | Version: Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Blocked By: Test Case: | Related Tickets: Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- Now that we have the capability of producing code with DWARF symbols, we should distribute GHC with them, or at least give the option. The DWARF symbols in these case would be both for the C RTS and for the base Haskell libraries shipped with GHC. Building GHC with said symbols amounts to adding {{{ GhcRtsHcOpts += -g GhcLibHcOpts += -g }}} to `mk/build.mk`. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10601> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #13870: Empty record construction for record-less constructor gives surprising runtime error (and surprisingly few warnings)
by GHC 16 Sep '17

16 Sep '17

#13870: Empty record construction for record-less constructor gives surprising runtime error (and surprisingly few warnings) -------------------------------------+------------------------------------- Reporter: RyanGlScott | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 Keywords: newcomer | Operating System: Unknown/Multiple Architecture: | Type of failure: Poor/confusing Unknown/Multiple | error message Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- The crux of this ticket is this sort of code: {{{#!hs module Main where f :: Maybe Int f = Just{} main :: IO () main = print f }}} {{{ $ runghc --ghc-arg=-Wall Bug.hs Just Bug.hs: Bug.hs:4:5-10: Missing field in record construction }}} Yikes. There are a couple of very surprising things happening here. First, the message `Missing field in record construction` is very misleading. After all, `Just` has no records! We really should give a more specific error which highlights this fact. (The fact that you can even use record construction syntax with a record- less constructor in the first place is a bit baffling, but the [https://www.haskell.org/onlinereport/haskell2010/haskellch3.html#x8-520003.… language standard does allow it], so I suppose we have to live with this.) The second surprising bit is the fact that is program compiled with no warnings whatsoever. This is in contrast to, say, the `Identity` datatype, which has a record (`runIdentity`): {{{#!hs module Main where import Data.Functor.Identity f :: Identity Int f = Identity{} main :: IO () main = print f }}} {{{ $ runghc --ghc-arg=-Wall Bug.hs Bug.hs:6:5: warning: [-Wmissing-fields] • Fields of ‘Identity’ not initialised: runIdentity • In the expression: Identity {} In an equation for ‘f’: f = Identity {} Identity Bug.hs: Bug.hs:6:5-14: Missing field in record construction runIdentity }}} Here, GHC warned me that I was doing something stupid. GHC ought to be warning me with equivalent fervor when I use `Just{}`. The warning would obviously have to be tweaked a bit, since warning that `Fields of ‘Just’ not initialised` doesn't make any sense, but you get the idea. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13870> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #12700: Don't warn about redundant constraints for type equalities
by GHC 15 Sep '17

15 Sep '17

#12700: Don't warn about redundant constraints for type equalities -------------------------------------+------------------------------------- Reporter: crockeea | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Incorrect Unknown/Multiple | warning at compile-time Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- With `-Wall`, the following code {{{ {-# LANGUAGE TypeFamilies #-} type family Foo x where Foo Double = Int foo :: (RealFrac a, Integral b, b ~ Foo a) => a -> b foo = round }}} produces the warning {{{ • Redundant constraint: b ~ Foo a • In the type signature for: foo :: (RealFrac a, Integral b, b ~ Foo a) => a -> b }}} Technically, I suppose this is correct in the sense that if we remove the constraint `b ~ Foo a`, `foo` still compiles. However, the ''meaning'' changes without the type equality, because the function I wrote is equivalent to `foo :: (RealFrac a, Integral (Foo a)) => a -> Foo a` (which requires `-XFlexibleContexts`), while the function it suggests is the much more general `foo :: (RealFrac a, Integral b) => a -> b`. Since this is the case, I think the warning is invalid. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12700> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #11649: LLVM code generator produces mal-formed LLVM blocks
by GHC 15 Sep '17

15 Sep '17

#11649: LLVM code generator produces mal-formed LLVM blocks -------------------------------------+------------------------------------- Reporter: bgamari | Owner: Type: bug | Status: new Priority: highest | Milestone: 8.0.1 Component: Compiler | Version: 8.0.1-rc2 (CodeGen) | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Compile-time Unknown/Multiple | crash Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- It appears that the recent addition (673efccb3b348e9daf23d9e65460691bbea8586e) of some instances for types in `GHC.Generics` has uncovered a bug in the LLVM code generator (at least on ARM). `libraries/base/GHC/Generics.hs` fails to compile with, {{{ Entry block to function must not have predecessors! label %cjGw /home/ben/ghc/roots/llvm-3.7/bin/opt: libraries/base/GHC/Generics.ll: error: input module is broken! }}} The problematic block in question appears to be, {{{ define internal ghccc void @rhSw_info$def(i32* noalias nocapture %Base_Arg, i32* noalias nocapture %Sp_Arg, i32* noal ias nocapture %Hp_Arg, i32 %R1_Arg, i32 %R2_Arg, i32 %R3_Arg, i32 %R4_Arg, i32 %SpLim_Arg) align 4 nounwind prefix <{ i32, i32, i32}><{i32 65541, i32 0, i32 15}> { clpH: br label %clpH } }}} Which arose from this Cmm, {{{ ==================== Post control-flow optimisations ==================== 2016-02-26 10:39:28.656744 UTC [section ""data" . lvl_rhSw_closure" { lvl_rhSw_closure: const lvl_rhSw_info; }, lvl_rhSw_entry() // [] { info_tbl: [(clpH, label: lvl_rhSw_info rep:HeapRep static { Fun {arity: 1 fun_type: ArgSpec 5} })] stack_info: arg_space: 0 updfr_space: Nothing } {offset clpH: goto clpH; // CmmBranch } }] }}} Which arose from this input Cmm, {{{ ==================== Cmm produced by new codegen ==================== 2016-02-26 10:39:28.611641 UTC [section ""data" . lvl_rhSw_closure" { lvl_rhSw_closure: const lvl_rhSw_info; }, lvl_rhSw_entry() // [R2] { info_tbl: [(clpD, label: lvl_rhSw_info rep:HeapRep static { Fun {arity: 1 fun_type: ArgSpec 5} })] stack_info: arg_space: 4 updfr_space: Just 4 } {offset clpD: _shWa::P32 = R2; // CmmAssign goto clpz; // CmmBranch clpz: if ((old + 0) - <highSp> < SpLim) goto clpE; else goto clpF; // CmmCondBranch clpE: R2 = _shWa::P32; // CmmAssign R1 = lvl_rhSw_closure; // CmmAssign call (stg_gc_fun)(R2, R1) args: 4, res: 0, upd: 4; // CmmCall clpF: goto clpy; // CmmBranch clpy: goto shWb; // CmmBranch shWb: goto clpH; // CmmBranch clpH: goto shWb; // CmmBranch } }] }}} Which arose from this STG, {{{#!hs lvl_rhSw :: forall a_a99i. GHC.Generics.U1 a_a99i -> GHC.Generics.U1 [a_a99i] [GblId, Arity=1, Caf=NoCafRefs, Str=DmdType <L,U>x, Unf=OtherCon []] = sat-only \r [eta_shWa] let-no-escape { some_v_shWb [Occ=LoopBreaker] :: GHC.Generics.U1 [a12_a99i] [LclId, Str=DmdType b] = NO_CCS[] \u [] some_v_shWb; } in some_v_shWb; }}} Which arose from this Core, {{{#!hs lvl_rhSw :: forall a_a99i. U1 a_a99i -> U1 [a_a99i] [GblId, Arity=1, Caf=NoCafRefs, Str=DmdType <L,U>x] lvl_rhSw = \ (@ a12_a99i) _ [Occ=Dead] -> letrec { some_v_a99l [Occ=LoopBreaker] :: U1 [a12_a99i] [LclId, Str=DmdType b] some_v_a99l = some_v_a99l; } in some_v_a99l }}} Which is apparently the body of the `some` implementation in the `Alternative` instance for `GHC.Generics.U1`. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11649> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 17