March 2018 - ghc-tickets

[GHC] #14959: Heep overflow in optimizer
by GHC 25 Mar '18

25 Mar '18

#14959: Heep overflow in optimizer -------------------------------------+------------------------------------- Reporter: darchon | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.4.1 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: -------------------------------------+------------------------------------- Compiling the following with optimisations: {{{ module Test where import Data.Bits (setBit) f = foldl setBit 0 [x | (x,_) <- zip [0..] [1]] :: Integer }}} fails with: {{{ $ ghc -O0 -fforce-recomp Test.hs [1 of 1] Compiling Test ( Test.hs, Test.o ) $ ghc -O -fforce-recomp Test.hs [1 of 1] Compiling Test ( Test.hs, Test.o ) ghc: panic! (the 'impossible' happened) (GHC version 8.4.1 for x86_64-unknown-linux): heap overflow }}} Fails on 8.0.2, 8.2.2, and 8.4.1 -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14959> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 9

Re: [GHC] #4442: Add unaligned version of indexWordArray#
by GHC 25 Mar '18

25 Mar '18

#4442: Add unaligned version of indexWordArray# -------------------------------------+------------------------------------- Reporter: tibbe | Owner: reinerp Type: feature request | Status: new Priority: low | Milestone: Component: Compiler | Version: 7.1 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: Type of failure: Runtime | Unknown/Multiple performance bug | Test Case: Blocked By: | Blocking: Related Tickets: #14447 | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by David Feuer <David.Feuer@…>): In [changeset:"efd70cfb4b0b9932a880ab417d75eaf95da3d5e6/ghc" efd70cfb/ghc]: {{{ #!CommitTicketReference repository="ghc" revision="efd70cfb4b0b9932a880ab417d75eaf95da3d5e6" Add unaligned bytearray access primops. Fixes #4442. Reviewers: bgamari, simonmar Reviewed By: bgamari Subscribers: dfeuer, rwbarton, thomie, carter GHC Trac Issues: #4442 Differential Revision: https://phabricator.haskell.org/D4488 }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/4442#comment:19> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #14925: Non-ASCII type names get garbled when their `TypeRep` is shown
by GHC 25 Mar '18

25 Mar '18

#14925: Non-ASCII type names get garbled when their `TypeRep` is shown -------------------------------------+------------------------------------- Reporter: | Owner: (none) leftaroundabout | Type: bug | Status: new Priority: low | Milestone: Component: | Version: 8.2.1 libraries/base | Keywords: Typeable | Operating System: Unknown/Multiple TypeRep Unicode ASCII UTF-8 | Architecture: | Type of failure: Incorrect result Unknown/Multiple | at runtime Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- [http://hackage.haskell.org/package/base-4.10.1.0/docs/Data-Typeable.html Typeable] allows easily showing the name of a type by, well, using `show` on it. However, this does not work right for types with Unicode symbols in their name: {{{ GHCi, version 8.2.1: http://www.haskell.org/ghc/ :? for help Prelude> :m +Data.Typeable Prelude Data.Typeable> data W = W Prelude Data.Typeable> typeOf W W Prelude Data.Typeable> data Ω = Ω Prelude Data.Typeable> typeOf Ω Î© }}} This did not yet happen in GHC-7: {{{ GHCi, version 7.10.2: http://www.haskell.org/ghc/ :? for help Prelude> :m +Data.Typeable Prelude Data.Typeable> data W = W Prelude Data.Typeable> typeOf W W Prelude Data.Typeable> data Ω = Ω Prelude Data.Typeable> typeOf Ω Ω }}} N.b.: {{{ Prelude> import Data.ByteString.Char8 as BC8 Prelude BC8> BC8.putStrLn $ pack "Î©" Ω }}} So, this appears to be a UTF-8 problem – something interprets bytestring- stored type-representation names as a different character encoding. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14925> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #14953: Panic when exporting duplicate record fields from separate modules
by GHC 25 Mar '18

25 Mar '18

#14953: Panic when exporting duplicate record fields from separate modules -------------------------------------+------------------------------------- Reporter: lyxia | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.5 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: -------------------------------------+------------------------------------- `A.hs` {{{ {-# LANGUAGE DuplicateRecordFields #-} module A where data R = R {unR :: Int} }}} --- `B.hs` {{{ {-# LANGUAGE DuplicateRecordFields #-} module B where data R = R {unR :: Int} }}} --- `C.hs` {{{ {-# LANGUAGE DuplicateRecordFields #-} module C (module A, module B) where import A import B }}} --- Output: {{{ C.hs:3:21: error:ghc-stage2: panic! (the 'impossible' happened) (GHC version 8.5.20180224 for x86_64-unknown-linux): exportClashErr $sel:unR:R Call stack: CallStack (from HasCallStack): callStackDoc, called at compiler/utils/Outputable.hs:1150:37 in ghc:Outputable pprPanic, called at compiler/typecheck/TcRnExports.hs:740:22 in ghc:TcRnExports }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14953> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #14916: Missing checks when deriving special classes
by GHC 25 Mar '18

25 Mar '18

#14916: Missing checks when deriving special classes -------------------------------------+------------------------------------- Reporter: kosmikus | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.2.2 (Type checker) | Keywords: | 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: -------------------------------------+------------------------------------- For the following program {{{ {-# LANGUAGE DeriveAnyClass #-} module T where import Data.Coerce import Data.Typeable data A = MkA deriving ((~) A) data B = MkB deriving (Coercible B) }}} the deriving clause for `A` is accepted without complaints, and the deriving clause for `B` fails with the following error: {{{ T.hs:8:24: error: Top-level bindings for unlifted types aren't allowed: | 8 | data B = MkB deriving (Coercible B) | ^^^^^^^^^^^ }}} Corresponding standalone deriving instances trigger errors saying "Manual instances of this class are not permitted". Probably similar error messages should be triggered here. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14916> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #14918: GHC 8.4.1 regression: derived Read instances with field names containing # no longer parse
by GHC 25 Mar '18

25 Mar '18

#14918: GHC 8.4.1 regression: derived Read instances with field names containing # no longer parse -------------------------------------+------------------------------------- Reporter: RyanGlScott | Owner: (none) Type: bug | Status: new Priority: high | Milestone: 8.4.2 Component: Compiler | Version: 8.4.1 Keywords: deriving | Operating System: Unknown/Multiple Architecture: | Type of failure: Incorrect result Unknown/Multiple | at runtime Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- (Originally noticed [here](https://github.com/ekmett/transformers- compat/issues/32).) Consider the following program: {{{#!hs {-# LANGUAGE MagicHash #-} module Bug where data T a = MkT { runT# :: a } deriving (Read, Show) t1, t2 :: T Int t1 = MkT 1 t2 = read $ show t1 main :: IO () main = print t2 }}} In GHC 8.2.1, this runs without issue: {{{ $ /opt/ghc/8.2.2/bin/runghc Bug.hs MkT {runT# = 1} }}} In GHC 8.4.1, however, this produces a runtime error: {{{ $ ~/Software/ghc-8.4.1/bin/runghc Bug.hs Bug.hs: Prelude.read: no parse }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14918> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #14885: TH breaks the scoping of quoted default method implementations when spliced
by GHC 25 Mar '18

25 Mar '18

#14885: TH breaks the scoping of quoted default method implementations when spliced -------------------------------------+------------------------------------- Reporter: RyanGlScott | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Template | Version: 8.2.2 Haskell | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: GHC rejects Unknown/Multiple | valid program Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Consider the following program: {{{#!hs {-# LANGUAGE ScopedTypeVariables #-} {-# LANGUAGE TemplateHaskell #-} {-# OPTIONS_GHC -ddump-splices #-} module Bug where class Foo1 a where bar1 :: forall b. a -> b -> b bar1 _ x = (x :: b) $([d| class Foo2 a where bar2 :: forall b. a -> b -> b bar2 _ x = (x :: b) |]) }}} `Foo1` typechecks, so naturally you'd expect `Foo2` to typecheck as well. Prepare to be surprised: {{{ $ /opt/ghc/8.2.2/bin/ghc Bug.hs [1 of 1] Compiling Bug ( Bug.hs, Bug.o ) Bug.hs:(10,3)-(13,6): Splicing declarations [d| class Foo2_aoA a_aoC where bar2_aoB :: forall b_aoD. a_aoC -> b_aoD -> b_aoD bar2_aoB _ x_aoE = (x_aoE :: b_aoD) |] ======> class Foo2_a3JQ a_a3JS where bar2_a3JR :: forall b_a3JT. a_a3JS -> b_a3JT -> b_a3JT bar2_a3JR _ x_a3JU = x_a3JU :: b_aoD Bug.hs:10:3: error: • Couldn't match expected type ‘b1’ with actual type ‘b’ ‘b’ is a rigid type variable bound by the type signature for: bar2 :: forall b. a0 -> b -> b at Bug.hs:(10,3)-(13,6) ‘b1’ is a rigid type variable bound by an expression type signature: forall b1. b1 at Bug.hs:(10,3)-(13,6) • In the expression: x_a3JU :: b In an equation for ‘bar2’: bar2 _ x_a3JU = x_a3JU :: b • Relevant bindings include x_a3JU :: b (bound at Bug.hs:10:3) bar2 :: a0 -> b -> b (bound at Bug.hs:10:3) | 10 | $([d| class Foo2 a where | ^^^^^^^^^^^^^^^^^^^^^^... }}} Notice how in the quoted `Foo2` declaration, the scoping is correct: `b_a0D` is used in both the type signature for `bar2_a0B` as well as in its default implementation. But after splicing, there are now two different `b`s: the one in the type signature (`b_a3JT`), and the one in the default implementation (`b_aoD`)! This causes the resulting type error. This is a regression that was introduced somewhere between 7.10.3 and 8.0.1, since it works in 7.10.3: {{{ $ /opt/ghc/7.10.3/bin/ghci Bug.hs GHCi, version 7.10.3: http://www.haskell.org/ghc/ :? for help [1 of 1] Compiling Bug ( Bug.hs, interpreted ) Bug.hs:(10,3)-(13,6): Splicing declarations [d| class Foo2_awn a_awp where bar2_awo :: forall b_awq. a_awp -> b_awq -> b_awq bar2_awo _ x_awr = (x_awr :: b_awq) |] ======> class Foo2_a3zs a_a3zu where bar2_a3zt :: forall b_awq. a_a3zu -> b_awq -> b_awq bar2_a3zt _ x_a3zv = x_a3zv :: b_awq Ok, modules loaded: Bug. }}} But not in any version of GHC since 8.0.1. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14885> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #14821: -O2 forces -optlo-O3 in a way that cannot be overridden at command line
by GHC 25 Mar '18

25 Mar '18

#14821: -O2 forces -optlo-O3 in a way that cannot be overridden at command line -------------------------------------+------------------------------------- Reporter: joeyhess | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.2 (LLVM) | 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 would expect -O2 -optlo-O2 to pass -O2 to the llvm optimiser. However, it seems that -O2 implies -optlo-O3, overriding the command-line's -optlo-O2. The workaround seems to be to use -O1 -optlo-O2, but of course this misses some ghc optimizations. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14821> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #14931: Segfault compiling files with -fprof-all
by GHC 25 Mar '18

25 Mar '18

#14931: Segfault compiling files with -fprof-all -------------------------------------+------------------------------------- Reporter: RyanGlScott | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Profiling | Version: 8.4.1 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: -------------------------------------+------------------------------------- Originally noticed [https://github.com/llvm-hs/llvm- hs/issues/86#issuecomment-373710312 here]. Take the following two files: {{{#!hs {-# LANGUAGE FlexibleInstances #-} {-# LANGUAGE FunctionalDependencies #-} {-# LANGUAGE MultiParamTypeClasses #-} {-# LANGUAGE UndecidableInstances #-} module State (MonadState(..), Lazy.evalState) where import qualified Control.Monad.Trans.State.Lazy as Lazy (StateT, get, put, evalState) class Monad m => MonadState s m | m -> s where get :: m s put :: s -> m () instance Monad m => MonadState s (Lazy.StateT s m) where get = Lazy.get put = Lazy.put }}} {{{#!hs {-# LANGUAGE FlexibleContexts #-} {-# LANGUAGE TemplateHaskell #-} module Bug where import Prelude (Int, IO, Bool(..), Num(..), Monad(..), not, print) import qualified Language.Haskell.TH.Syntax as TH import State wat :: IO () wat = print $(let playGame [] = do (_, score) <- get return score playGame (x:xs) = do (on, score) <- get case x of 'a' | on -> put (on, score + 1) 'b' | on -> put (on, score - 1) 'c' -> put (not on, score) _ -> put (on, score) playGame xs startState :: (Bool, Int) startState = (False, 0) in TH.lift (evalState (playGame "abcaaacbbcabbab") startState) ) }}} Compiling them like so leads to a segfault: {{{ $ ~/Software/ghc-8.4.1/bin/ghc -c -O -static -dynamic-too -dynosuf dyn_o -dynhisuf dyn_hi State.hs $ ~/Software/ghc-8.4.1/bin/ghc -c -O -prof -osuf p_o -hisuf p_hi State.hs -fprof-auto $ ~/Software/ghc-8.4.1/bin/ghc -c -O -prof -osuf p_o -hisuf p_hi Bug.hs Segmentation fault (core dumped) }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14931> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 13

[GHC] #13324: Allow PartialTypeSignatures in the instance context of a standalone deriving declaration
by GHC 23 Mar '18

23 Mar '18

#13324: Allow PartialTypeSignatures in the instance context of a standalone deriving declaration -------------------------------------+------------------------------------- Reporter: RyanGlScott | Owner: (none) Type: feature | Status: new request | Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 (Type checker) | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: #10607 Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Currently, if you try to use a wildcard anywhere in an instance context, it will fail immediately: {{{ $ ghci GHCi, version 8.0.2: http://www.haskell.org/ghc/ :? for help Loaded GHCi configuration from /home/rgscott/.ghci λ> instance _ => Show (Maybe a) <interactive>:1:10: error: Wildcard ‘_’ not allowed in an instance declaration for ‘Show’ }}} But there's one situation where we could lift this restriction: the context of a standalone, derived instance declaration. That is, something like this: {{{#!hs deriving instance _ => Show (Maybe a) }}} Why? Because GHC already has the machinery needed to infer what the context should be (see [http://git.haskell.org/ghc.git/blob/8a6b8c5fb96472d7c2e2cd1918950dd50f2fef7… this part] of `TcDerivInfer`), so if a user turned on `PartialTypeSignatures`, GHC could just fill in the wildcard with the inferred constraints. The implementation won't be //that// easy, however, since we'd also have to watch out for trickery such as: {{{#!hs deriving instance (C a, _) => C (T a b c) }}} I only mentioned putting wildcards in a derived instance //context//, because I think allowing the use of wildcards elsewhere in the instance head might be too difficult to deal with. I mean, how would you fill in this, for example? {{{#!hs instance (_ a) }}} This mini-feature has a very practical application: it would allow users to wield the flexibility of `StandaloneDeriving` without having to manually type in the instance context every time. That is, users could type in these instances: {{{#!hs deriving instance _ => Data (T a b c) deriving instance _ => Eq (T a b c) }}} Instead of their fully spelled-out, more laborious counterparts. This would be crucial for Template Haskell, as its ability to infer these contexts is quite limited (see #10607 for an example where this cropped up). Idle thought: could this be generalized to work for the instance context of //any// instance declaration (and not just derived ones)? From an outside perspective, it seems like typechecking other instances would require inferring constraints in a fashion quite similar to that of derived instances. But I am not at all familiar with that part of the typechecker, so I might be totally off here. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13324> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 12