- ghc-tickets - Haskell.org

Re: [GHC] #8013: Strange closure type error building hs-kqueue on FreeBSD
by GHC 16 Jul '14

16 Jul '14

#8013: Strange closure type error building hs-kqueue on FreeBSD -------------------------------------+------------------------------------- Reporter: ahktenzero | Owner: Type: bug | Status: infoneeded Priority: normal | Milestone: 7.10.1 Component: Compiler | Version: 7.6.3 (LLVM) | Keywords: Resolution: | Operating System: FreeBSD Differential Revisions: | Type of failure: Compile-time Architecture: x86_64 | crash (amd64) | Test Case: Difficulty: Unknown | Blocking: Blocked By: | Related Tickets: | -------------------------------------+------------------------------------- Comment (by ahktenzero): You can close this if you want, I reset the options for GHC and all the Haskell based ports on my system to the defaults a while back and haven't had a reoccurrence of the problem since. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8013#comment:5> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #7828: RebindableSyntax and Arrow
by GHC 16 Jul '14

16 Jul '14

#7828: RebindableSyntax and Arrow -------------------------------------+------------------------------------- Reporter: | Owner: jstolarek AlessandroVermeulen | Status: new Type: bug | Milestone: 7.10.1 Priority: normal | Version: 7.6.2 Component: Compiler | Keywords: (Type checker) | Operating System: Unknown/Multiple Resolution: | Type of failure: GHC rejects Differential Revisions: | valid program Architecture: | Test Case: Unknown/Multiple | Blocking: Difficulty: Unknown | Blocked By: | Related Tickets: #1537, | #3613 | -------------------------------------+------------------------------------- Comment (by ross): Replying to [comment:42 jstolarek]: > I'm able to compile > > {{{ > test :: Arrow a => a i i > test = proc n -> do > returnA -< n > }}} > > using the new desugaring, which was not possible earlier. I wouldn't have expected any change in the handling of this. > But I'm still stuck on this: > > {{{ > test :: Arrow a => a i i > test = proc n -> do > returnA -< n > returnA -< n > }}} If we're allowing rebinding, that should be type-checked and desugared exactly as if it were written {{{ test :: Arrow a => a i i test = proc n -> (returnA -< n) `thenA` (returnA -< n) }}} Here's the type inference (backwards, the way the checker does it): {{{ D |- proc n -> (returnA -< n) `thenA` (returnA -< n) :: a i t ---------------------------------------------------------------- (ProcExpr) D; n::i |-a (returnA -< n) `thenA` (returnA -< n) : () --> t ------------------------------------------------------------ (ArrForm) D |- thenA :: forall e. a1 (e,s1) i -> a2 (e,s2) i -> a (e,()) t D; n::i |-a1 returnA -< n : s1 --> i ------------------------------------ (ArrApp) D |- returnA :: a1 i i D, n::i |- n :: i D; n::i |-a2 returnA -< n : s2 --> i ------------------------------------ (ArrApp) D |- returnA :: a2 i i D, n::i |- n :: i }}} For the default `thenA`, {{{ thenA :: Arrow a => a (e,s) b -> a (e,s) c -> a (e,s) c u `thenA` v = arr id &&& u >>> arr fst >>> v }}} we have a1 = a2 = a, s1 = s2 = () and t = i, but with rebinding they'd be taken from the type of `thenA`, which could be anything matching the type in the above inference. > I noticed that desugaring of `cmd` stored in the `BodyStmt` always passes `()` as the type of the stack. That's true for the default, but if we have rebinding it will be determined by the type of `thenA`/`bind`. > Ross, I also tried to implement desugaring of `bind`. When generating the desugared Core for `cmd 'bind' \ p -> do { ss }` I have problem with the `\p ->` part. So, given the desugared `do { ss }`, `cmd` and the `bind` operator how should the generated command lambda look like? I'm still at a loss to undestand how to explicitly manipulate the stack from within Core. Here's the type inference: {{{ D; xs |-a cmd1 `bind` \ p -> cmd2 : s --> t ------------------------------------------- (ArrForm) D |- bind :: forall e. a1 (e,s1) t1 -> a2 (e,(b,s2)) t1 -> a (e,s) t D; xs |-a1 :: cmd1 : s1 --> t1 D; xs |-a2 :: \ p -> cmd2 : (b,s2) --> t2 ----------------------------------------- (Lam) D; xs, p::b |-a2 cmd2 : s2 --> t2 }}} In this case the manipulation of the stack is done by the translation of `CmdLam` and by the code inside the vanilla Haskell implementation of `bind`. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7828#comment:43> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #8910: cross compiling for x86_64 solaris2
by GHC 16 Jul '14

16 Jul '14

#8910: cross compiling for x86_64 solaris2 ----------------------------------+---------------------------------- Reporter: maeder | Owner: Type: task | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.1-rc2 Keywords: | Operating System: Solaris Architecture: x86_64 (amd64) | Type of failure: Runtime crash Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ----------------------------------+---------------------------------- using a "gcc -m64" and a working ghc-7.8.0.20140228-i386-unknown-solaris2 I was able to create a 64Bit ghc-stage2 compiler that can compile a simple hello world program but may create binaries from larger sources that seg- fault. In particular "ghc-stage2 --interactive" seg-faults: {{{ [Thread debugging using libthread_db enabled] [New Thread 1 (LWP 1)] [New LWP 2 ] [New LWP 3 ] [New LWP 4 ] GHCi, version 7.8.0.20140228: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... linking ... done. Loading package integer-simple ... linking ... done. Loading package base ... linking ... Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 1 (LWP 1)] 0x00000000041ea4c0 in ?? () (gdb) bt #0 0x00000000041ea4c0 in ?? () #1 0x000000000369cf92 in resolveObjs () #2 0x000000000206de99 in cplU_info () #3 0x000000000000000c in ?? () #4 0x0000000003d0fe09 in base_GHCziEventziManager_zdLr88clvl8_closure () #5 0x0000000000000000 in ?? () }}} The content of my mk/build.mk is just {{{ INTEGER_LIBRARY = integer-simple }}} Creating a binary-dist also worked, but unpacking and "gmake install" failed at first because old-time, haskell98 and haskell2010 libraries were missing due to "!CrossCompiling" in ghc.mk and the second time with: {{{ Reading package info from "rts/dist/package.conf.install" ... done. "utils/ghc-cabal/dist-install/build/tmp/ghc-cabal-bindist" register libraries/ghc-prim dist-install "/local/home/maeder/haskell/ghc-7.8-x64-static/lib/ghc-7.8.0.20140228/bin/ghc" "/local/home/maeder/haskell/ghc-7.8-x64-static/lib/ghc-7.8.0.20140228/bin /ghc-pkg" "/local/home/maeder/haskell/ghc-7.8-x64-static/lib/ghc-7.8.0.20140228" '' '/local/home/maeder/haskell/ghc-7.8-x64-static' '/local/home/maeder/haskell/ghc-7.8-x64-static/lib/ghc-7.8.0.20140228' '/local/home/maeder/haskell/ghc-7.8-x64-static/share/doc/ghc/html/libraries' NO ghc-cabal: Bad interface file: dist-install/build/GHC/CString.hi magic number mismatch: old/corrupt interface file? (wanted 129742, got 33214052) gmake[1]: *** [install_packages] Fehler 1 }}} What am I doing wrong for cross compiling? There is also #8378 using a different build host. #8713 also seem to use x86_64 solaris2 libs. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8910> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 12

[GHC] #9288: Type class overlapping instances check doesn't understand type equality
by GHC 16 Jul '14

16 Jul '14

#9288: Type class overlapping instances check doesn't understand type equality -------------------------------------+------------------------------------- Reporter: ezyang | Owner: Type: bug | Status: new Priority: low | Milestone: Component: Compiler (Type | Version: 7.9 checker) | Operating System: Unknown/Multiple Keywords: | Type of failure: GHC accepts Architecture: Unknown/Multiple | invalid program Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | -------------------------------------+------------------------------------- Here is a sample program which we'd expect to give an overlapping instances error, but does not: {{{ {-# LANGUAGE TypeFamilies, FlexibleInstances #-} module Foo where type family F a :: * type instance F Bool = Int type instance F Int = Bool class C a where instance (F Bool ~ a) => C a where instance C Int where }}} Fortunately, GHC does notice when you actually try to use the instance. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9288> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

Re: [GHC] #7936: newStdGen leaks memory when result is not used
by GHC 15 Jul '14

15 Jul '14

#7936: newStdGen leaks memory when result is not used -------------------------------------+------------------------------------- Reporter: | Owner: ekmett ryantrinkle | Status: patch Type: bug | Milestone: 7.10.1 Priority: normal | Version: 7.6.3 Component: | Keywords: libraries/random | Operating System: Unknown/Multiple Resolution: | Type of failure: Runtime crash Differential Revisions: | Test Case: Architecture: | Blocking: Unknown/Multiple | Difficulty: Unknown | Blocked By: | Related Tickets: | -------------------------------------+------------------------------------- Changes (by simonpj): * cc: ekmett (added) * owner: rrnewton => ekmett Comment: Making Edward K owner, since it's a Core Libraries issue Simon -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7936#comment:4> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #4218: System.Random is way too lazy
by GHC 15 Jul '14

15 Jul '14

#4218: System.Random is way too lazy -------------------------------------+------------------------------------- Reporter: EyalLotem | Owner: ekmett Type: bug | Status: patch Priority: low | Milestone: 7.10.1 Component: | Version: 6.12.3 libraries/random | Keywords: random stack Resolution: | overflow Differential Revisions: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: Difficulty: Unknown | Blocking: Blocked By: | Related Tickets: | -------------------------------------+------------------------------------- Changes (by simonpj): * cc: ekmett (added) * owner: rrnewton => ekmett Comment: Making Edward K owner, since it's a Core Libraries issue. Just a question of reviewing the offered patch! Simon -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/4218#comment:12> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #5051: Typechecker behaviour change
by GHC 15 Jul '14

15 Jul '14

#5051: Typechecker behaviour change -------------------------------------+------------------------------------- Reporter: igloo | Owner: Type: bug | Status: new Priority: normal | Milestone: 7.10.1 Component: Compiler | Version: 7.0.2 Resolution: | Keywords: Differential Revisions: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: Difficulty: Unknown | typecheck/should_compile/T5051 Blocked By: | Blocking: Related Tickets: | -------------------------------------+------------------------------------- Comment (by simonpj): As I say in comment:5, I think this is the right behaviour, so I don't propose to fix it at all. Only leaving it open in case others trip over something similar. Simon -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/5051#comment:8> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #9299: GHCi type inference error
by GHC 15 Jul '14

15 Jul '14

#9299: GHCi type inference error ------------------------------------+------------------------------------- Reporter: erikd | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: GHCi | 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: | ------------------------------------+------------------------------------- This code (required conduit-1.1.6): {{{ {- Bolled down from https://github.com/snoyberg/conduit/blob/process/conduit- extra/Data/Conduit/Process.hs -} {-# LANGUAGE FlexibleInstances #-} {-# LANGUAGE TypeFamilies #-} {-# LANGUAGE RankNTypes #-} import System.Process import Control.Monad.IO.Class (MonadIO, liftIO) import System.IO (Handle, hClose) import Data.Conduit import Data.ByteString (ByteString) import Data.Maybe (fromMaybe) import Control.Applicative ((<$>), (<*>)) data UseProvidedHandle = UseProvidedHandle data Inherited = Inherited data ClosedStream = ClosedStream class InputSource a where isStdStream :: (Maybe Handle -> IO a, Maybe StdStream) instance InputSource Handle where isStdStream = (\(Just h) -> return h, Just CreatePipe) instance InputSource ClosedStream where isStdStream = (\(Just h) -> hClose h >> return ClosedStream, Just CreatePipe) instance (r ~ (), MonadIO m, i ~ ByteString) => InputSource (ConduitM i o m r) where isStdStream = (\(Just h) -> return $ sinkHandle h, Just CreatePipe) instance (r ~ (), r' ~ (), MonadIO m, MonadIO n, i ~ ByteString) => InputSource (ConduitM i o m r, n r') where isStdStream = (\(Just h) -> return (sinkHandle h, liftIO $ hClose h), Just CreatePipe) instance InputSource Inherited where isStdStream = (\Nothing -> return Inherited, Just Inherit) instance InputSource UseProvidedHandle where isStdStream = (\Nothing -> return UseProvidedHandle, Nothing) sinkHandle :: MonadIO m => Handle -> Consumer ByteString m () sinkHandle = error "sinkHandle" conduitProcess :: (MonadIO m, InputSource stdin) => CreateProcess -> m (stdin, ProcessHandle) conduitProcess cp = liftIO $ do let (getStdin, stdinStream) = isStdStream (stdinH, _, _, ph) <- createProcess cp { std_in = fromMaybe (std_in cp) stdinStream } (,) <$> getStdin stdinH <*> return ph main :: IO () main = putStrLn "Hello" }}} Compiles fine with ghc, and runs correctly with `runghc`, but fails to load into ghci with the following error: {{{ GHCi, version 7.8.3: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... done. [1 of 1] Compiling Main ( test.hs, interpreted ) test.hs:45:9: No instance for (InputSource a0) arising from the ambiguity check for ‘stdinStream’ The type variable ‘a0’ is ambiguous When checking that ‘stdinStream’ has the inferred type ‘Maybe StdStream’ Probable cause: the inferred type is ambiguous In the second argument of ‘($)’, namely ‘do { let (getStdin, stdinStream) = isStdStream; (stdinH, _, _, ph) <- createProcess (cp {std_in = fromMaybe (std_in cp) stdinStream}); (,) <$> getStdin stdinH <*> return ph }’ In the expression: liftIO $ do { let (getStdin, stdinStream) = isStdStream; (stdinH, _, _, ph) <- createProcess (cp {std_in = fromMaybe (std_in cp) stdinStream}); (,) <$> getStdin stdinH <*> return ph } Failed, modules loaded: none. }}} Same problem in 7.6.3, 7.8.2 and 7.8.3. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9299> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 9

[GHC] #9263: Iface type variable out of scope with default associated types and polykinds
by GHC 15 Jul '14

15 Jul '14

#9263: Iface type variable out of scope with default associated types and polykinds ------------------------------------+------------------------------------- Reporter: goldfire | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.9 Keywords: | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- I have {{{ {-# LANGUAGE DataKinds, PolyKinds, TypeFamilies #-} module A where import Data.Proxy class kproxy ~ 'KProxy => PEq (kproxy :: KProxy a) where type F (x :: a) :: Bool type F (x :: a) = False }}} {{{ {-# LANGUAGE DataKinds, KindSignatures, TypeFamilies #-} module B where import A import Data.Proxy data Void instance PEq ('KProxy :: KProxy Void) }}} {{{ module C where import B }}} Now, this happens in a terminal window: {{{ rae:21:38:44 ~/temp> ghc A.hs [1 of 1] Compiling A ( A.hs, A.o ) rae:21:38:46 ~/temp> ghc B.hs [2 of 2] Compiling B ( B.hs, B.o ) rae:21:38:47 ~/temp> ghc C.hs [3 of 3] Compiling C ( C.hs, C.o ) The interface for ‘B’ Declaration for R:FVoidx: Iface type variable out of scope: a Cannot continue after interface file error }}} If any two of the files are compiled together, the error does not occur. The emptiness of the `Void` type is irrelevant -- the error initially occurred with an inhabited type. This is a regression from 7.8.2. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9263> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #9264: Scoped kind variables do not work with default associated types
by GHC 15 Jul '14

15 Jul '14

#9264: Scoped kind variables do not work with default associated types ------------------------------------+------------------------------------- Reporter: goldfire | 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: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- I have {{{ {-# LANGUAGE PolyKinds, TypeFamilies, ScopedTypeVariables #-} module Bug where class C (a :: k) where type F (a :: k) type F (a :: k) = Int }}} Compiling gives me {{{ Bug.hs:7:11: The signature specified kind ‘k1’, but ‘a’ has kind ‘k2’ In the type ‘(a :: k)’ In the type instance declaration for ‘F’ In the class declaration for ‘C’ }}} The error is on the line declaring the default associated type for `F`. This happens in both 7.8.2 and HEAD. Having !ScopedTypeVariables causes no change, but I put it in the example to emphasize that adding !ScopedTypeVariables is not the solution. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9264> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3