May 2016 - ghc-tickets

[GHC] #12098: Typechecker regression in 8.0.1
by GHC 23 May '16

23 May '16

#12098: Typechecker regression in 8.0.1 -------------------------------------+------------------------------------- Reporter: vagarenko | Owner: Type: bug | Status: new Priority: highest | Milestone: Component: Compiler | Version: 8.0.1 (Type checker) | Keywords: | Operating System: Windows Architecture: x86 | Type of failure: GHC rejects | valid program Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- This code: {{{#!hs {-# LANGUAGE IncoherentInstances #-} {-# LANGUAGE RankNTypes #-} {-# LANGUAGE RoleAnnotations #-} module Bug where import Data.Proxy (Proxy(..)) import GHC.Prim (coerce) class Throws e where {} type role Throws representational newtype Wrap e a = Wrap { unWrap :: Throws e => a } coerceWrap :: Wrap e a -> Wrap (Catch e) a coerceWrap = coerce newtype Catch a = Catch a instance Throws (Catch e) where {} unthrow :: Proxy e -> (Throws e => a) -> a unthrow _ = unWrap . coerceWrap . Wrap }}} compiles fine with ghc 7.10.2 but fails with ghc 8.0.1 with error: {{{ Bug.hs:25:13: error: * Could not deduce (Throws e) from the context: Throws e0 bound by a type expected by the context: Throws e0 => a at Bug.hs:25:13-38 Possible fix: add (Throws e) to the context of the type signature for: unthrow :: Proxy e -> (Throws e => a) -> a * In the expression: unWrap . coerceWrap . Wrap In an equation for `unthrow': unthrow _ = unWrap . coerceWrap . Wrap }}} This code is extracted from blog post http://www.well- typed.com/blog/2015/07/checked-exceptions/ and gist https://gist.github.com/edsko/f1f566f77422398fba7d -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12098> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

[GHC] #12101: Regression: Pattern synonyms make GHCi 8.0.1 crash
by GHC 22 May '16

22 May '16

#12101: Regression: Pattern synonyms make GHCi 8.0.1 crash -------------------------------------+------------------------------------- Reporter: int-index | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: GHCi | Version: 8.0.1 Keywords: | Operating System: Unknown/Multiple PatternSynonyms | Architecture: | Type of failure: GHCi crash Unknown/Multiple | Test Case: | Blocked By: https://gist.githubusercontent.com | /int- | index/17dfb68fd97f724aef2849d0defae0d6/raw/3dab71756bbafa197b93170310ccabfea8c92120/M.hs| Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- This code: {{{ {-# LANGUAGE PatternSynonyms #-} data T = C pattern P :: T pattern P = C pattern P' :: T pattern P' = P }}} compiles fine with both GHC 7.10.3 and 8.0.1. However, if you try to load it into GHCi 8.0.1, it panics: {{{ GHCi, version 8.0.1: http://www.haskell.org/ghc/ :? for help Prelude> :l M.hs [1 of 1] Compiling Main ( M.hs, interpreted ) ghc: panic! (the 'impossible' happened) (GHC version 8.0.1 for x86_64-unknown-linux): kindPrimRep.go rep_a18a Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug }}} GHCi 7.10.3 does not crash, therefore it's a regression: {{{ GHCi, version 7.10.3: http://www.haskell.org/ghc/ :? for help Prelude> :l M.hs [1 of 1] Compiling Main ( M.hs, interpreted ) Ok, modules loaded: Main. *Main> }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12101> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

[GHC] #12068: RULE too complicated to desugar when using constraint synonyms
by GHC 22 May '16

22 May '16

#12068: RULE too complicated to desugar when using constraint synonyms -------------------------------------+------------------------------------- Reporter: crockeea | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.3 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 the following minimal-as-I-could-get-it example, {{{ {-# LANGUAGE ConstraintKinds #-} class Qux a class (Num r) => Class1 r class (Num r) => Class2 r newtype Foo q z = Foo z type Qux' q z = (Qux q, Integral z) instance (Num z) => Num (Foo q z) instance (Qux' q z, Num z) => Class1 (Foo q z) instance (Qux' q z, Num z) => Class2 (Foo q z) newtype Bar r = Bar r {-# SPECIALIZE bar :: (Qux q) => Bar (Foo q Int) -> Bar (Foo q Int) #-} bar :: (Class1 r, Class2 r) => Bar r -> Bar r bar = undefined }}} I get the warning: {{{ RULE left-hand side too complicated to desugar Optimised lhs: let { $dNum_aGE :: Num (Foo q Int) [LclId, Str=DmdType] $dNum_aGE = Main.$fNumFoo @ q @ Int GHC.Num.$fNumInt } in bar @ (Foo q Int) (Main.$fClass1Foo @ q @ Int $dNum_aGE ($dQux_aFH, $dIntegral_aGI) GHC.Num.$fNumInt) (Main.$fClass2Foo @ q @ Int $dNum_aGE ($dQux_aFH, $dIntegral_aGI) GHC.Num.$fNumInt) Orig lhs: let { $dIntegral_aGI :: Integral Int [LclId, Str=DmdType] $dIntegral_aGI = GHC.Real.$fIntegralInt } in let { tup_aGJ :: Qux' q Int [LclId, Str=DmdType] tup_aGJ = ($dQux_aFH, $dIntegral_aGI) } in let { $dNum_aGH :: Num Int [LclId, Str=DmdType] $dNum_aGH = GHC.Num.$fNumInt } in let { $dNum_aGG :: Num Int [LclId, Str=DmdType] $dNum_aGG = $dNum_aGH } in let { tup_aGF :: Qux' q Int [LclId, Str=DmdType] tup_aGF = ($dQux_aFH, $dIntegral_aGI) } in let { $dNum_aGE :: Num (Foo q Int) [LclId, Str=DmdType] $dNum_aGE = Main.$fNumFoo @ q @ Int $dNum_aGH } in let { $dClass2_aFK :: Class2 (Foo q Int) [LclId, Str=DmdType] $dClass2_aFK = Main.$fClass2Foo @ q @ Int $dNum_aGE tup_aGJ $dNum_aGH } in let { $dClass1_aFJ :: Class1 (Foo q Int) [LclId, Str=DmdType] $dClass1_aFJ = Main.$fClass1Foo @ q @ Int $dNum_aGE tup_aGF $dNum_aGG } in bar @ (Foo q Int) $dClass1_aFJ $dClass2_aFK }}} This is apparently due to my use of a constraint synonym. In this code it would be quite simple to just replace the synonym with the constraints on its RHS, but in my real code my constraint synonym is an associated type, so that is not an option. It would be great o be able to specialize in the presence of constraint synonyms. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12068> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

Re: [GHC] #1853: hpc mix files for Main modules overwrite each other
by GHC 22 May '16

22 May '16

#1853: hpc mix files for Main modules overwrite each other ----------------------------------+-------------------------------------- Reporter: guest | Owner: Type: bug | Status: new Priority: lowest | Milestone: Component: Code Coverage | Version: 6.8.1 Resolution: | Keywords: Operating System: Linux | Architecture: x86_64 (amd64) Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | ----------------------------------+-------------------------------------- Changes (by thomie): * cc: kostmo (added) Comment: Perhaps relevant to the discussion here, @kostmo points out in #12065: > specifying `HPCTIXFILE` as an environment variable when running the executable program does change the output path of the .tix file. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/1853#comment:28> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #9095: make sdist picks up test files
by GHC 22 May '16

22 May '16

#9095: make sdist picks up test files ------------------------------------+------------------------------------- Reporter: ezyang | Owner: Type: bug | Status: new Priority: low | Milestone: Component: Build System | 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 recently ran 'make sdist' and discovered that I my source tarball was nearly a gigabyte large. Closer inspection revealed that sdist had slurped up executable files for tests that I had run at some point. It would be preferable if 'make sdist' did not do that. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9095> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #11574: schedule: re-entered unsafely on minimal hello world program on arm
by GHC 22 May '16

22 May '16

#11574: schedule: re-entered unsafely on minimal hello world program on arm -------------------------------------+------------------------------------- Reporter: andrewufrank | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.3 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: -------------------------------------+------------------------------------- i installed ghc 7.10.3 from jessie-backports (together with the corresponding LLVM 1:3.5.2.3~bpo8+2) and get for the minimal main = putStrLn "hello" program the error message {{{testthree: schedule: re-entered unsafely. Perhaps a 'foreign import unsafe' should be 'safe'? }} i am not certain if this is related: cabal install for the same minimal program (with a minimal cabal file) hangs. cabal install -v gives {{{ /home/frank/.cabal/setup-exe-cache/setup-Simple-Cabal-1.22.5.0-arm-linux- ghc-7.10.3 configure --verbose=2 --ghc --prefix=/home/frank/.cabal --bindir=/home/frank/.cabal/bin --libdir=/home/frank/.cabal/lib --libsubdir=arm-linux-ghc-7.10.3/testthree-0.0.6-3AgAf9bDkWJGZdbGLvO875 --libexecdir=/home/frank/.cabal/libexec --datadir=/home/frank/.cabal/share --datasubdir=arm-linux-ghc-7.10.3/testthree-0.0.6 --docdir=/home/frank/.cabal/share/doc/arm-linux-ghc-7.10.3/testthree-0.0.6 --htmldir=/home/frank/.cabal/share/doc/arm-linux- ghc-7.10.3/testthree-0.0.6/html --haddockdir=/home/frank/.cabal/share/doc/arm-linux- ghc-7.10.3/testthree-0.0.6/html --sysconfdir=/home/frank/.cabal/etc --user --extra-prog-path=/home/frank/.cabal/bin --dependency=base=base-4.8.2.0-2f1f71a7fcf013cd47fd21f489064f9a --disable-tests --exact-configuration --disable-benchmarks Redirecting build log to {handle: /home/frank/.cabal/logs/testthree-0.0.6.log} }}} any comment how this can be fixed? thank you - a running 7.10.3 for armhf is a great achievement! -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11574> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 10

Re: [GHC] #4370: Bring back monad comprehensions
by GHC 22 May '16

22 May '16

#4370: Bring back monad comprehensions -------------------------------------+------------------------------------- Reporter: simonpj | Owner: Type: feature request | Status: closed Priority: normal | Milestone: ⊥ Component: Compiler | Version: 6.12.3 Resolution: fixed | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Changes (by bgamari): * status: patch => closed * differential: Phab:D2247 => * resolution: => fixed Comment: Of dear, indeed it isn't; sorry for the noise. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/4370#comment:72> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #6077: Respect XDG_CONFIG_HOME
by GHC 22 May '16

22 May '16

#6077: Respect XDG_CONFIG_HOME -------------------------------------+------------------------------------- Reporter: So8res | Owner: Type: feature request | Status: new Priority: normal | Milestone: Component: None | Version: 7.4.1 Resolution: | Keywords: Operating System: Linux | Architecture: | Unknown/Multiple Type of failure: Other | Test Case: Blocked By: 5966 | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by siddhanathan): The GHCi configuration file would clearly be in `$XDG_CONFIG_HOME/ghc/ghci.conf`. Where exactly the GHCi history should be placed is debatable. [https://wiki.debian.org/XDGBaseDirectorySpecification Debian] argues for an additional state directory. But for now, I think `$XDG_DATA_HOME/ghc/ghci_history` would be best. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/6077#comment:5> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #10961: Make it possible to purely use the parser
by GHC 22 May '16

22 May '16

#10961: Make it possible to purely use the parser -------------------------------------+------------------------------------- Reporter: mpickering | Owner: Type: task | Status: new Priority: low | Milestone: Component: Compiler | Version: 7.10.2 (Parser) | Keywords: newcomer | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- It is claimed that the parser is pure but one of the arguments is `DynFlags`. The only safe way to construct a copy of `DynFlags` requires IO so it is not possible to use the parser purely. I propose to introduce a new datatype `ParserOptions` which contains the options used by the parser and a function `DynFlags -> ParserOptions` so that it is possible to purely construct a `ParserOptions` and use the parser purely. This would be useful for external users of the parser. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10961> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 9

[GHC] Batch modify: #10547
by GHC 21 May '16

21 May '16

Batch modification to #10547 by bgamari: milestone to 8.0.2 Comment: Ticket retargeted after milestone closed -- Tickets URL: <http://ghc.haskell.org/trac/ghc/query?id=10547> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0