- ghc-tickets - Haskell.org

Re: [GHC] #5546: Documentation errors in Control.Exception.Base
by GHC 05 May '17

05 May '17

#5546: Documentation errors in Control.Exception.Base -------------------------------------+------------------------------------- Reporter: bit | Owner: (none) Type: bug | Status: new Priority: high | Milestone: ⊥ Component: libraries/base | Version: 8.0.1 Resolution: | 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 bezirg): * owner: simonmar => (none) * status: closed => new * version: 7.2.1 => 8.0.1 * resolution: fixed => * milestone: 7.4.1 => ⊥ -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/5546#comment:4> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #5546: Documentation errors in Control.Exception.Base
by GHC 05 May '17

05 May '17

#5546: Documentation errors in Control.Exception.Base -------------------------------------+------------------------------------- Reporter: bit | Owner: simonmar Type: bug | Status: closed Priority: high | Milestone: 7.4.1 Component: libraries/base | Version: 7.2.1 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: | -------------------------------------+------------------------------------- Comment (by bezirg): In the documentation of [https://hackage.haskell.org/package/base-4.9.1.0/docs/Control- Exception.html Control.Exception] of base 4.9.10 there is still a link to non-existing ''forkIOUnmasked'' function: > To create a a new thread in an unmasked state use forkIOUnmasked. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/5546#comment:3> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #11799: Legend for hpc markup colors
by GHC 05 May '17

05 May '17

#11799: Legend for hpc markup colors -------------------------------------+------------------------------------- Reporter: drathier | Owner: Type: feature | Status: new request | Priority: normal | Milestone: Component: Test Suite | 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 hpc markup colors are not documented anywhere. Coming from an imperative background, red means never executed and green means executed at least once. Here, green means always true, red means always false, yellow means never executed. Adding a one-line legend at the top of the output file(s) describing the colours would help a lot. Suggesting `never executed` `always true` and `always false` on a single line at the top with the same yellow, green and red background as the respective highlighted code would have. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11799> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #11616: Kinds aren't instantiated
by GHC 05 May '17

05 May '17

#11616: Kinds aren't instantiated -------------------------------------+------------------------------------- Reporter: Iceland_jack | Owner: Type: task | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.1 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: -------------------------------------+------------------------------------- {{{#!hs {-# LANGUAGE TypeInType, AllowAmbiugousTypes #-} class Whoami a where whoami :: String instance Whoami Int where whoami = "int" instance Whoami Bool where whoami = "[y/n]" instance Whoami Maybe where whoami = "call me maybe" }}} we can write {{{ >>> whoami @Type @Int "int" >>> whoami @Type @Bool "[y/n]" >>> whoami @(Type -> Type) @Maybe "call me maybe" }}} Attempting to specialise `whoami` to `Type` we can't simply write `whoamiType = whoami @Type` as one might expect. To start we define a synonym: {{{ $ ghci -ignore-dot-ghci /tmp/t17q.hs GHCi, version 8.1.20160117: http://www.haskell.org/ghc/ :? for help [1 of 1] Compiling Main ( /tmp/t17q.hs, interpreted ) Ok, modules loaded: Main. *Main> whoami2 = whoami <interactive>:1:1: error: • Ambiguous type variable ‘a0’ arising from a use of ‘whoami’ prevents the constraint ‘(Whoami a0)’ from being solved. Probable fix: use a type annotation to specify what ‘k0’, ‘a0’ should be. These potential instances exist: three instance involving out-of-scope typess (use -fprint-potential-instances to see them all) • When instantiating ‘whoami2’, initially inferred to have this overly-general type: forall k (a :: k). Whoami a => String NB: This instantiation can be caused by the monomorphism restriction. }}} Using the “overly-general” inferred type `whoami :: forall k (a :: k). Whoami a => String` fails: {{{#!hs -- • Could not deduce (Whoami a0) arising from a use of ‘whoami’ -- from the context: Whoami a -- bound by the type signature for: -- whoami2 :: Whoami a => String -- at /tmp/t17q.hs:20:1-44 -- The type variable ‘a0’ is ambiguous -- These potential instances exist: -- three instance involving out-of-scope typess -- (use -fprint-potential-instances to see them all) -- • In the expression: whoami -- In an equation for ‘whoami2’: whoami2 = whoami whoami2 :: forall k (a :: k). Whoami a => String whoami2 = whoami }}} so we need `TypeApplications`: {{{#!hs whoami2 :: forall k (a :: k). Whoami a => String whoami2 = whoami @_ @a }}} and then we can write: {{{#!hs whoamiType :: forall (a :: Type). Whoami a => String whoamiType = whoami @_ @a }}} and we can write as intended: {{{ >>> whoamiType @Int "int" >>> whoamiType @Bool "[y/n]" }}} Is there a reason one cannot write: {{{#!hs whoami2 = whoami whoamiType = whoami @Type }}} ---- As a side note, what is the difference between {{{#!hs -- >>> :set -fprintf-explicit-foralls -- >>> :kind Whoami1 -- Whoami1 :: forall {k}. k -> Constraint -- >>> :kind Whoami2 -- Whoami2 :: forall k. k -> Constraint class Whoami1 a class Whoami2 (a :: k) }}} Is there a snippet of code that highlights differences -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11616> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 8

[GHC] #13632: Frontend plugin arguments are reversed
by GHC 05 May '17

05 May '17

#13632: Frontend plugin arguments are reversed -------------------------------------+------------------------------------- Reporter: duog | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: GHC API | Version: 8.0.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: frontend01 | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- The arguments passed to a frontend plugin are passed in the opposite order to that with which they were passed to the command line. I will attach a diff that modifies the frontend01 test to exhibit the error. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13632> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #13320: Unfortunate compiler loop when creating type loop (with UndecidableInstances)
by GHC 05 May '17

05 May '17

#13320: Unfortunate compiler loop when creating type loop (with UndecidableInstances) -------------------------------------+------------------------------------- Reporter: Ptival | Owner: (none) Type: bug | Status: new Priority: low | Milestone: Component: Compiler | Version: 8.0.1 Keywords: | Operating System: Unknown/Multiple UndecidableInstances loop | Architecture: | Type of failure: Compile-time Unknown/Multiple | crash or panic Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- I'm afraid this will simply be seen as "that's what happens when you use UndecidableInstances", but I might as well document this issue I had. Trying to play with a "Trees that Grow" syntax, I encountered an issue when making a mistake, that can be boiled down to the following: {{{#!hs {-# language ConstraintKinds, FlexibleContexts, TypeFamilies, UndecidableInstances #-} module Loop where import GHC.Exts (Constraint) import Test.QuickCheck type family X_Var ξ data TermX ξ = Var (X_Var ξ) type ForallX (φ :: * -> Constraint) ξ = ( φ (X_Var ξ) ) --genTerm :: ForallX Arbitrary ξ => Int -> Gen (TermX ξ) genTerm 0 = Var <$> arbitrary genTerm n = Var <$> genTerm (n - 1) --instance ForallX Arbitrary ξ => Arbitrary (TermX ξ) where --arbitrary = sized genTerm }}} This code will compile correctly, and generate: {{{#!hs genTerm :: (X_Var ξ ~ TermX ξ, Arbitrary (TermX ξ), Eq t, Num t) => t -> Gen (TermX ξ) }}} Which is correct (though, not the type I had intended, since my code had a mistake). Now, if you uncomment the "instance" line only, the compiler will loop. Adding the commented out type, of course, gives a type error where it's due. I was just wondering whether this type of error failed with the loops that should be caught by the compiler. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13320> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 8

[GHC] #13642: GHCi 8.2 simply ignores TH splice using datatype with a forall'd kind signature
by GHC 05 May '17

05 May '17

#13642: GHCi 8.2 simply ignores TH splice using datatype with a forall'd kind signature -------------------------------------+------------------------------------- Reporter: RyanGlScott | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Template | Version: 8.2.1-rc1 Haskell | 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: -------------------------------------+------------------------------------- This is a pretty bizarre behavior I've noticed recently that only happens in GHC 8.2 or later. If you try to use a Template Haskell splice with a very particular feature (a datatype whose kind uses `forall`) in GHCi, then GHCi will flat-out ignore it! {{{ $ /opt/ghc/8.2.1/bin/ghci GHCi, version 8.2.0.20170427: http://www.haskell.org/ghc/ :? for help Loaded GHCi configuration from /home/rgscott/.ghci λ> :set -XGADTs -XTypeInType -XTemplateHaskell -XRankNTypes λ> import Language.Haskell.TH (stringE, pprint) λ> import Data.Kind (Type) λ> $([d| data Foo :: forall a. a -> Type where MkFoo :: Foo Int |] >>= stringE . pprint) λ> print (5 + length $([d| data Foo :: forall a. a -> Type where MkFoo :: Foo Int |] >>= stringE . pprint)) λ> 5 5 λ> it 5 λ> $([d| data Foo :: forall a. a -> Type where MkFoo :: Foo Int |] >>= stringE . pprint) λ> it 5 }}} Notice how none of my attempts to use the splice seemed to register with GHCi. This isn't really a regression //per se//, since GHC 8.0.1 nor 8.0.2 even allowed you to get that far: {{{ $ /opt/ghc/8.0.2/bin/ghci GHCi, version 8.0.2: http://www.haskell.org/ghc/ :? for help Loaded GHCi configuration from /home/rgscott/.ghci λ> :set -XGADTs -XTypeInType -XTemplateHaskell -XRankNTypes λ> import Language.Haskell.TH (stringE, pprint) λ> import Data.Kind (Type) λ> $([d| data Foo :: forall a. a -> Type where MkFoo :: Foo Int |] >>= stringE . pprint) <interactive>:4:3: error: Exotic form of kind not (yet) handled by Template Haskell forall a. a -> Type }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13642> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #13645: whoCreated produces an uninformative stack trace when an exception is raised in a CAF
by GHC 04 May '17

04 May '17

#13645: whoCreated produces an uninformative stack trace when an exception is raised in a CAF -------------------------------------+------------------------------------- Reporter: refold | Owner: (none) 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: -------------------------------------+------------------------------------- Consider the following program: {{{#!hs {-# LANGUAGE ScopedTypeVariables #-} module Main where import Control.Exception import GHC.Stack {-# NOINLINE caf #-} caf :: [Int] caf = [1..500] {-# NOINLINE caf_exc #-} caf_exc :: Int caf_exc = caf !! 1000 {-# NOINLINE foo #-} foo :: Int -> Int foo 0 = caf_exc foo n = foo $ n - 1 {-# NOINLINE bar #-} bar :: Int -> Int bar n = bar' n where bar' 0 = foo n bar' m = bar' $ m - 1 main :: IO () main = print (bar 10) `catch` \(e :: SomeException) -> do stacktrace <- whoCreated e print stacktrace }}} By default, when built with profiling, `whoCreated` in the example above produces a quite uninformative stack trace: {{{#!shell $ ./caf-nostack ["GHC.List.CAF (<entire-module>)"] }}} However, if you run the program with `+RTS -xc`, you'll see that it prints a stack trace with much more context: {{{#!shell $ ./caf-nostack +RTS -xc *** Exception (reporting due to +RTS -xc): (THUNK_2_0), stack trace: GHC.List.CAF --> evaluated by: Main.caf_exc, called from Main.CAF --> evaluated by: Main.foo, called from Main.bar.bar', called from Main.bar, called from Main.main, called from Main.CAF --> evaluated by: Main.main ["GHC.List.CAF (<entire-module>)"] }}} It'd be nice if `whoCreated` produced something closer to the `+RTS -xc` output in this case. Cabalised test project: https://github.com/23Skidoo/caf-nostack -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13645> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #13638: configure error: GHC is required.
by GHC 04 May '17

04 May '17

#13638: configure error: GHC is required. -------------------------------------+------------------------------------- Reporter: QuRyu | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Build System | Version: 8.0.1 Keywords: | Operating System: MacOS X Architecture: | Type of failure: Building GHC Unknown/Multiple | failed Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- While I was trying to download GHC source code from Github and to build it myself, I encountered the error "configure: error: GHC is required." when I was doing "./configure". The "perl boot" raised no warning or error, and the source code was directly downloaded from Github using "git clone --recursive git://github.com/ghc/ghc.git". -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13638> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #13631: In GHCi a result is wrong when -fdefer-typed-holes is used with underscore alone
by GHC 03 May '17

03 May '17

#13631: In GHCi a result is wrong when -fdefer-typed-holes is used with underscore alone -------------------------------------+------------------------------------- Reporter: vanto | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.2 Keywords: | 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: -------------------------------------+------------------------------------- I open this ticket instead of {{{#13579}}} to restate it and I closed the other.\\ In fact a result of the compiler which I think is wrong misled me.\\ Therefore I close tickets {{{#13602}}} and {{{#13557}}}.\\ In Haskell2010 Language Report it is said that :\\ 1. underscore "_ " all by itself is a reserved identifier.\\ 2. underscore "_" is treated as a lowercase letter, and can occur wherever a lowercase letter can.\\ So {{{_e}}} is an identifier like {{{__}}}\\ GHCi gives a bad result when he computed the code below.\\ {{{ Prelude> :set -fdefer-typed-holes Prelude> let f = map (\x -> True) [_, _] <interactive>:2:27: warning: [-Wtyped-holes] * Found hole: _ :: a0 Where: `a0' is an ambiguous type variable * In the expression: _ In the second argument of `map', namely `[_, _]' In the expression: map (\ x -> True) [_, _] * Relevant bindings include f :: [Bool] (bound at <interactive>:2:5) <interactive>:2:30: warning: [-Wtyped-holes] * Found hole: _ :: a0 Where: `a0' is an ambiguous type variable * In the expression: _ In the second argument of `map', namely `[_, _]' In the expression: map (\ x -> True) [_, _] * Relevant bindings include f :: [Bool] (bound at <interactive>:2:5) Prelude> f [True,True] }}} The underscore "_" is alone , all by itself and yet is recognized as an identifier, and GHCi gives a result.\\ This is the bug. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13631> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1