- ghc-tickets - Haskell.org

[GHC] #12835: ghc: panic! (the 'impossible' happened) - find_tycon
by GHC 14 Nov '16

14 Nov '16

#12835: ghc: panic! (the 'impossible' happened) - find_tycon -------------------------------------+------------------------------------- Reporter: mrijkeboer | Owner: Type: bug | Status: new Priority: high | Milestone: Component: Compiler | Version: 8.0.1 Keywords: | Operating System: Linux Architecture: x86_64 | Type of failure: Compile-time (amd64) | crash or panic Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- {{{ $ stack build server-0.0.0.1: build -- While building package server-0.0.0.1 using: /home/martijn/.stack/setup-exe-cache/x86_64-linux/setup-Simple- Cabal-1.24.0.0-ghc-8.0.1 --builddir=.stack- work/dist/x86_64-linux/Cabal-1.24.0.0 build lib:server exe:server --ghc- options " -ddump-hi -ddump-to-file" Process exited with code: ExitFailure 1 Logs have been written to: /home/martijn/server/.stack- work/logs/server-0.0.0.1.log Preprocessing library server-0.0.0.1... [20 of 20] Compiling Application ( src/Application.hs, .stack- work/dist/x86_64-linux/Cabal-1.24.0.0/build/Application.o ) ghc: panic! (the 'impossible' happened) (GHC version 8.0.1 for x86_64-unknown-linux): find_tycon YesodSubRunnerEnv [] Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12835> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

[GHC] #12834: GHC panic: while printing Non type-variable argument
by GHC 14 Nov '16

14 Nov '16

#12834: GHC panic: while printing Non type-variable argument -------------------------------------+------------------------------------- Reporter: phadej | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 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: -------------------------------------+------------------------------------- {{{ {-# LANGUAGE GADTs, TypeFamilies, DataKinds, TypeOperators, MultiParamTypeClasses, UndecidableInstances, UndecidableSuperClasses, FlexibleInstances, PolyKinds, KindSignatures #-} import GHC.Exts (Constraint) newtype I a = I a data NP :: (k -> *) -> [k] -> * where Nil :: NP f '[] (:*) :: f x -> NP f xs -> NP f (x ': xs) infixr 5 :* class (AllF f xs, SListI xs) => All (f :: k -> Constraint) (xs :: [k]) instance (AllF f xs, SListI xs) => All f xs data SList :: [k] -> * where SNil :: SList '[] SCons :: SListI xs => SList (x ': xs) class SListI (xs :: [k]) where -- | Get hold of the explicit singleton (that one can then -- pattern match on). sList :: SList xs instance SListI '[] where sList = SNil instance SListI xs => SListI (x ': xs) where sList = SCons -- | Type family used to implement 'All'. -- type family AllF (c :: k -> Constraint) (xs :: [k]) :: Constraint type instance AllF _c '[] = () type instance AllF c (x ': xs) = (c x, All c xs) semigroup :: All ((~) (Maybe Int)) xs => NP I xs -> NP I xs -> NP I xs semigroup = undefined }}} Causes {{{ ghc-failure-all.hs:37:14: error: • Non type-variable argumentghc: panic! (the 'impossible' happened) (GHC version 8.0.1 for x86_64-apple-darwin): print_equality ~ }}} If `AllF` is used directly in the definition of `sappend`, there is no error whatsoever. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12834> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #11660: Remove Type pretty-printer in favor of IfaceType
by GHC 14 Nov '16

14 Nov '16

#11660: Remove Type pretty-printer in favor of IfaceType -------------------------------------+------------------------------------- Reporter: bgamari | Owner: Type: task | 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: -------------------------------------+------------------------------------- The `Type` pretty-printer has a fair amount of code in common with the pretty-printer for `IfaceType`. Moreover, both cases handle a number of special cases. With the introduction of `-fprint-explicit-runtime-reps` (Phab:D1961), there is now an inconsistency between these two. Let's resolve this inconsistency by eliminating the `Type` pretty-printer in favor of using the `IfaceType` printer. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11660> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 20

[GHC] #12831: Fulltext search SQL error in Trac
by GHC 12 Nov '16

12 Nov '16

#12831: Fulltext search SQL error in Trac -------------------------------------+------------------------------------- Reporter: skovalev | Owner: hvr Type: bug | Status: new Priority: normal | Milestone: Component: Trac & Git | Version: 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 have faced with `Trac` errors while searching next queries: * `ghc: panic!` * `the 'impossible' happened` * `Report execution failed` -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12831> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

[GHC] #12816: Link error with GOLD linker
by GHC 12 Nov '16

12 Nov '16

#12816: Link error with GOLD linker -------------------------------------+------------------------------------- Reporter: hsyl20 | Owner: Type: bug | Status: new Priority: high | Milestone: 8.0.2 Component: Compiler | Version: 8.1 (Linking) | Keywords: | Operating System: Linux Architecture: x86_64 | Type of failure: Building GHC (amd64) | failed Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- 1. I think we should add "pthread" to the list of extra-libs in "rts/package.conf.in" for the threaded RTS on UNIX-like host OSes. 2. we should include MachDeps.h in "rts/package.conf.in" Current master branch (2e8463b232054b788b73e6551947a9434aa76009) build is broken on my system[0]. When ghc-stage1 tries to produce ghc-stage2, I get: {{{ /home/hsyl20/repo/ghc/rts/dist/build/libHSrts_thr-ghc8.1.20161107.so: error: undefined reference to 'pthread_create' /home/hsyl20/repo/ghc/rts/dist/build/libHSrts_thr-ghc8.1.20161107.so: error: undefined reference to 'pthread_detach }}} (logs: http://pastebin.com/EbqEx6Gg ) I have tracked down the regression to the following recent commit: a977c96537bb7077c6445f02db98636b150e6e14 If I revert this commit, it builds. However I think this commit has only revealed the bug: if I add "pthread" to the list of extra-libs in "rts/package.conf.in", it builds too. We already add it on freebsd and openbsd but not on Linux. I think I have finally found out why other devs on #ghc haven't noticed this bug: my system uses the GOLD linker (because of #12748), but if I switch back to the BFD linker, it builds without error. While investigating this bug, I have noticed that linker flags for 64-bit atomic ops introduced by https://git.haskell.org/ghc.git/commitdiff/c06e3f46d24ef69f3a3d794f5f604cb8… aren't set while they should: WORD_SIZE_IN_BITS isn't defined because we don't include MachDeps.h [0] x86-64, ArchLinux (Linux 4.8.6), GNU gold (GNU Binutils 2.27) 1.12, gcc (GCC) 6.2.1 20160830 -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12816> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #12812: Debugging functions should throw warnings
by GHC 12 Nov '16

12 Nov '16

#12812: Debugging functions should throw warnings -------------------------------------+------------------------------------- Reporter: siddhanathan | Owner: Type: task | Status: new Priority: lowest | Milestone: Component: | Version: 8.0.1 libraries/base | 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: -------------------------------------+------------------------------------- The function Debug.Trace.trace is useful for debugging, however, we don't want this function to be present in production code. An excerpt from Debug.Trace explains why: {{{ The 'trace' function should /only/ be used for debugging, or for monitoring execution. The function is not referentially transparent: its type indicates that it is a pure function but it has the side effect of outputting the trace message. }}} It would be useful if the compiler threw warnings when such functions are used. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12812> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

[GHC] #12829: Multiline input (‘:set +m’) terminated by trailing whitespace
by GHC 12 Nov '16

12 Nov '16

#12829: Multiline input (‘:set +m’) terminated by trailing whitespace -------------------------------------+------------------------------------- Reporter: Iceland_jack | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: GHCi | Version: 8.0.1 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: -------------------------------------+------------------------------------- {{{ $ ghci -ignore-dot-ghci GHCi, version 8.0.1: http://www.haskell.org/ghc/ :? for help Prelude> :set +m Prelude> let Prelude| a :: Int Prelude| a = 5 Prelude| Prelude> }}} works fine with [https://downloads.haskell.org/~ghc/master/users- guide/ghci.html#multiline-input multiline input], but if you add one trailing space it immediately terminates the input {{{ Prelude> let Prelude| a :: Int␣ <interactive>:11:1: error: The type signature for ‘a’ lacks an accompanying binding }}} I don't know if this is the desired behavior but this means I can't copy/paste code into GHCi that has a trailing whitespace. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12829> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #7679: Regression in -fregs-graph performance
by GHC 12 Nov '16

12 Nov '16

#7679: Regression in -fregs-graph performance -------------------------------------+------------------------------------- Reporter: simonmar | Owner: Type: bug | Status: new Priority: low | Milestone: Component: Compiler (NCG) | Version: 7.6.2 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: Type of failure: Runtime | Unknown/Multiple performance bug | Test Case: Blocked By: | Blocking: Related Tickets: #7192, #8657 | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Changes (by bgamari): * milestone: 8.0.2 => Comment: This isn't an immediate priority (although tjakway has expressed interest in picking it up). -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7679#comment:15> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #12021: Type variable escapes its scope
by GHC 12 Nov '16

12 Nov '16

#12021: Type variable escapes its scope -------------------------------------+------------------------------------- Reporter: ttuegel | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1-rc3 Keywords: | Operating System: Linux Architecture: x86_64 | Type of failure: GHC rejects (amd64) | valid program Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- In the following example, I believe a type variable is incorrectly escaping its scope. The example produces the error: {{{ Bug.hs:20:20: error: • Expected kind ‘Cat k’, but ‘(:-)’ has kind ‘Constraint -> Constraint -> *’ • In the first argument of ‘Iso’, namely ‘(:-)’ In the type signature: functor :: forall a b. Iso (:-) (:-) (Ob (Dom f) a) (Ob (Dom f) b) (Ob (Cod f) (f a)) (Ob (Cod f) (f b)) In the class declaration for ‘Functor’ }}} If the definition of `Iso` is changed from {{{#!hs type Iso (c :: Cat k) (d :: Cat k) s t a b = forall p. (Cod p ~ Nat d (->)) => p a b -> p s t }}} to {{{#!hs type Iso (c :: Cat k) (d :: Cat k) s t a b = forall p. p a b -> p s t }}} then this error does not occur. (The example still will not compile because I have omitted almost the entire implementation, but it should not fail here.) I am not certain what is really happening here, but it seems to me that when the RHS of `Iso` is constrained, then the type variable `k` introduced on the LHS of `Iso` is being unified incorrectly with the type variable `k` introduced in the definition of the `Functor` class. When the constraint is removed, GHC seems to recognize (correctly!) that the type variables are distinct. (This bug actually occurs with GHC 8.0.1-rc4, but the "Version" menu doesn't give me that option.) {{{#!hs {-# LANGUAGE ConstraintKinds #-} {-# LANGUAGE MultiParamTypeClasses #-} {-# LANGUAGE NoImplicitPrelude #-} {-# LANGUAGE RankNTypes #-} {-# LANGUAGE TypeFamilies #-} {-# LANGUAGE TypeInType #-} {-# LANGUAGE TypeOperators #-} import GHC.Base ( Constraint, Type ) type Cat k = k -> k -> Type class Category (p :: Cat k) where type Ob p :: k -> Constraint class (Category (Dom f), Category (Cod f)) => Functor (f :: j -> k) where type Dom f :: Cat j type Cod f :: Cat k functor :: forall a b. Iso (:-) (:-) (Ob (Dom f) a) (Ob (Dom f) b) (Ob (Cod f) (f a)) (Ob (Cod f) (f b)) class (Functor f , Dom f ~ p, Cod f ~ q) => Fun (p :: Cat j) (q :: Cat k) (f :: j -> k) | f -> p q instance (Functor f , Dom f ~ p, Cod f ~ q) => Fun (p :: Cat j) (q :: Cat k) (f :: j -> k) data Nat (p :: Cat j) (q :: Cat k) (f :: j -> k) (g :: j -> k) type Iso (c :: Cat k) (d :: Cat k) s t a b = forall p. (Cod p ~ Nat d (->)) => p a b -> p s t data (p :: Constraint) :- (q :: Constraint) }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12021> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #12797: Default Rules stop working when providing some constraints
by GHC 12 Nov '16

12 Nov '16

#12797: Default Rules stop working when providing some constraints -------------------------------------+------------------------------------- Reporter: danilo2 | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 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: -------------------------------------+------------------------------------- I've just found a very strange behavior. Let's consider following program: {{{ {-# LANGUAGE NoMonomorphismRestriction #-} {-# LANGUAGE TypeFamilies #-} {-# LANGUAGE ExtendedDefaultRules #-} {-# LANGUAGE OverloadedStrings #-} module Main where import Prelude import Control.Monad.IO.Class type family FuncArg m where FuncArg ((->) t) = 'Just t FuncArg m = 'Nothing test1 :: (MonadIO m) => m () test1 = do liftIO $ print "tst" test2 :: (MonadIO m, FuncArg m ~ 'Nothing) => m () test2 = do liftIO $ print "tst" main :: IO () main = return () }}} The function `tst1` compiles fine, while `tst2` fails: {{{ exe/Main.hs:21:14: error: • Could not deduce (Show a0) arising from a use of ‘print’ from the context: (MonadIO m, FuncArg m ~ 'Nothing) bound by the type signature for: tst2 :: (MonadIO m, FuncArg m ~ 'Nothing) => m () at exe/Main.hs:19:1-49 The type variable ‘a0’ is ambiguous These potential instances exist: instance Show Ordering -- Defined in ‘GHC.Show’ instance Show Integer -- Defined in ‘GHC.Show’ instance Show a => Show (Maybe a) -- Defined in ‘GHC.Show’ ...plus 22 others ...plus 7 instances involving out-of-scope types (use -fprint-potential-instances to see them all) • In the second argument of ‘($)’, namely ‘print "tst"’ In a stmt of a 'do' block: liftIO $ print "tst" In the expression: do { liftIO $ print "tst" } exe/Main.hs:21:20: error: • Could not deduce (Data.String.IsString a0) arising from the literal ‘"tst"’ from the context: (MonadIO m, FuncArg m ~ 'Nothing) bound by the type signature for: tst2 :: (MonadIO m, FuncArg m ~ 'Nothing) => m () at exe/Main.hs:19:1-49 The type variable ‘a0’ is ambiguous These potential instances exist: instance a ~ Char => Data.String.IsString [a] -- Defined in ‘Data.String’ ...plus one instance involving out-of-scope types (use -fprint-potential-instances to see them all) • In the first argument of ‘print’, namely ‘"tst"’ In the second argument of ‘($)’, namely ‘print "tst"’ In a stmt of a 'do' block: liftIO $ print "tst" }}} Giving explicit types to String literals fixes the problem. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12797> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 8