April 2016 - ghc-tickets

[GHC] #10547: feature request: expanding type synonyms in error messages
by GHC 26 Aug '16

26 Aug '16

#10547: feature request: expanding type synonyms in error messages -------------------------------------+------------------------------------- Reporter: osa1 | Owner: osa1 Type: feature | Status: new request | Milestone: 7.12.1 Priority: low | Version: Component: Compiler | Operating System: Unknown/Multiple (Type checker) | Type of failure: None/Unknown Keywords: | Blocked By: Architecture: | Related Tickets: Unknown/Multiple | Test Case: | Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- Type errors sometimes becoming confusing because of synonyms, for example, expected type and found type sometimes use different synonyms or sometimes only one of them use a synonym etc. We discussed this in the mailing list: https://mail.haskell.org/pipermail /ghc-devs/2015-June/009247.html And also moved it to the Wiki to collect examples: https://wiki.haskell.org/Expanding_type_synonyms_in_error_messages_proposal I'm currently working on a patch, I made some progress and I'll hopefully submit something for reviews in a couple of days. This ticket is for further discussions and keep track of the progress. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10547> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 23

[GHC] #11442: Segfault when showing (undefined :: Type)
by GHC 26 Aug '16

26 Aug '16

#11442: Segfault when showing (undefined :: Type) -------------------------------------+------------------------------------- Reporter: Iceland_jack | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.1 Keywords: | Operating System: Linux Architecture: x86_64 | Type of failure: Compile-time (amd64) | crash Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- This [https://ghc.haskell.org/trac/ghc/ticket/11311?replyto=3#comment:3 comment] made me wonder about the relationship between `Void` and `Type`, `id :: Void -> Void` and `id :: Type -> Type`: https://ghc.haskell.org/trac/ghc/ticket/11311?replyto=3#comment:3 {{{#!hs {-# LANGUAGE TypeSynonymInstances, FlexibleInstances #- import Data.Kind (Type) instance Show Type where show _ = "..." main = print (undefined :: Type) }}} running gives: {{{ $ runghc --version runghc 8.1.20160113 $ runghc -ignore-dot-ghci Segfault.hs Segmentation fault (core dumped) }}} Verbose log attached. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11442> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #11443: SPECIALIZE pragma does not work + performance drop in GHC 8.0-rc1
by GHC 25 Aug '16

25 Aug '16

#11443: SPECIALIZE pragma does not work + performance drop in GHC 8.0-rc1 -------------------------------------+------------------------------------- Reporter: danilo2 | Owner: Type: bug | Status: new Priority: high | Milestone: Component: Compiler | Version: 8.0.1-rc1 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: -------------------------------------+------------------------------------- Hello! I've just hit a strange issue. I might missinterpret how the `SPECIALIZE` pragma works, but if I understand correctly, then there is a bug in GHC. Lets consider this simple code: module `A`: {{{ {-# LANGUAGE UndecidableInstances #-} {-# LANGUAGE TypeOperators #-} {-# LANGUAGE KindSignatures #-} {-# LANGUAGE DataKinds #-} {-# LANGUAGE TypeFamilies #-} {-# LANGUAGE FlexibleInstances #-} module A where import Prelude import GHC.TypeLits -- TF utils type family (a :: Nat) :== (b :: Nat) where a :== a = 'True a :== b = 'False type family If cond (a :: Nat) (b :: Nat) where If 'True a b = a If 'False a b = b -- Heavy TF computations type family HeavyTF (n :: Nat) (i :: Nat) :: Nat where HeavyTF n 0 = 0 HeavyTF n i = If (HeavyTF' n :== 0) (HeavyTF n (i - 1)) 1 type family HeavyTF' (n :: Nat) :: Nat where HeavyTF' 0 = 0 HeavyTF' n = HeavyTF' (n - 1) -- Params for tests (bigger numbers = longer compile times) type family NatOf a :: Nat type instance NatOf Int = 12000 type instance NatOf String = 12000 -- Type class to check GHC behavior class PerfC1 a where perfc1 :: a -> String instance CheckOk (HeavyTF 10 (NatOf a)) => PerfC1 a where perfc1 _ = "oh" ; {-# INLINABLE perfc1 #-} class CheckOk (n :: Nat) instance CheckOk 0 -- where main_cache :: IO () main_cache = do print $ perfc1 (1 :: Int) print $ perfc1 ("a" :: String) perfc1_Int :: Int -> String perfc1_Int = perfc1 perfc1_String :: String -> String perfc1_String = perfc1 {-# SPECIALIZE perfc1 :: Int -> String #-} {-# SPECIALIZE perfc1 :: String -> String #-} ----- perfc1' :: PerfC1 a => a -> String perfc1' = perfc1 {-# INLINABLE perfc1' #-} {-# SPECIALIZE perfc1' :: Int -> String #-} {-# SPECIALIZE perfc1' :: String -> String #-} }}} module `Test1`: {{{ import A main = do print $ perfc1 (1 :: Int) print $ perfc1 ("a" :: String) }}} module `Test2`: {{{ import A main = do print $ perfc1' (1 :: Int) print $ perfc1' ("a" :: String) }}} module `Test3`: {{{ import A main = do print $ perfc1_Int (1 :: Int) print $ perfc1_String ("a" :: String) }}} Compile with: `ghc 7.10.3` : `ghc -O2 -fenable-rewrite-rules Test<n>.hs` `ghc 8.0-rc1` : `ghc -O2 -fenable-rewrite-rules -freduction-depth=0 Test<n>.hs` If module `A` was already compiled the compilation times for `ghc 7.10.3` were as follow: - `Test1`: ~ 16s - `Test2`: ~ 16s - `Test3`: almost instant And for `ghc 8.0-rc1` were as follow: - `Test1`: ~ 28s - `Test2`: ~ 28s - `Test3`: almost instant Here are 2 bugs to note: 1) the compilation times are much longer with new GHC 2) the specialize pragmas do not work -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11443> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 17

[GHC] #11632: Data.Char repeated readLitChar barfs on output from show "ó1"
by GHC 25 Aug '16

25 Aug '16

#11632: Data.Char repeated readLitChar barfs on output from show "ó1" -------------------------------------+------------------------------------- Reporter: inversemot | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: | Version: 7.10.3 libraries/base | Keywords: | Operating System: Linux Architecture: | Type of failure: Incorrect result Unknown/Multiple | at runtime Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- "ó1" is "\243\&1" and when shown that's "\"\\243\\&1\"" {{{#!hs readLitChar "\"\\243\\&1\"" = [("\"", "\243\\&1")] readLitChar "\243\\&1" = [("\243", "\\&1")] --should have consumed "\\&" readLitChar "\\&1" = [] }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11632> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 16

[GHC] #11960: GHC parallel build failure during "make"
by GHC 25 Aug '16

25 Aug '16

#11960: GHC parallel build failure during "make" -------------------------------------+------------------------------------- Reporter: ilovezfs | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Build System | Version: 8.0.1-rc3 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Building GHC Unknown/Multiple | failed Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- This keeps happening (in the same place every time): {{{ config.status: executing libtool commands config.status: executing include commands config.status: executing src commands # wc on OS X has spaces in its output, which libffi's Makefile # doesn't expect, so we tweak it to sed them out mv libffi/build/Makefile libffi/build/Makefile.orig sed "s#wc -w#wc -w | sed 's/ //g'#" < libffi/build/Makefile.orig > libffi/build/Makefile "touch" libffi/stamp.ffi.static-shared.configure make: *** [all] Error 2 }}} Adding some sleep in various places seems to work around it, so it appears to be a race of some kind during parallel builds. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11960> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 10

[GHC] #11825: Pretty printer doesn't display functional dependency with ‘→’
by GHC 25 Aug '16

25 Aug '16

#11825: Pretty printer doesn't display functional dependency with ‘→’ -------------------------------------+------------------------------------- Reporter: Iceland_jack | Owner: Type: feature | Status: new request | Priority: normal | Milestone: Component: GHCi | 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: -------------------------------------+------------------------------------- {{{ % ghci -ignore-dot-ghci GHCi, version 8.1.20160117: http://www.haskell.org/ghc/ :? for help Prelude> :set -XUnicodeSyntax Prelude> import Control.Monad.Free Prelude Control.Monad.Free> :set -XUnicodeSyntax Prelude Control.Monad.Free> :i MonadFree class Monad m => MonadFree (f :: * -> *) (m :: * -> *) | m -> f where wrap :: f (m a) -> m a default wrap :: (m ~ t n, Control.Monad.Trans.Class.MonadTrans t, MonadFree f n, Functor f) => f (m a) -> m a -- Defined in ‘Control.Monad.Free.Class’ instance [safe] Functor f => MonadFree f (Free f) -- Defined in ‘Control.Monad.Free’ Prelude Control.Monad.Free> }}} should be {{{ class Monad m => MonadFree (f :: * -> *) (m :: * -> *) | m → f where }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11825> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #11833: GHC can't derive polykinded instance of polykinded typeclass for newtype that requires a class constraint
by GHC 25 Aug '16

25 Aug '16

#11833: GHC can't derive polykinded instance of polykinded typeclass for newtype that requires a class constraint -------------------------------------+------------------------------------- Reporter: RyanGlScott | Owner: RyanGlScott Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 (Type checker) | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: GHC rejects Unknown/Multiple | valid program Test Case: | Blocked By: Blocking: | Related Tickets: #8865 Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- GHC chokes when trying to derive the following: {{{#!hs {-# LANGUAGE GeneralizedNewtypeDeriving #-} {-# LANGUAGE PolyKinds #-} module Example where class Category (cat :: k -> k -> *) where catId :: cat a a catComp :: cat b c -> cat a b -> cat a c newtype T (c :: * -> * -> *) a b = MkT (c a b) deriving Category }}} with the following error: {{{ $ /opt/ghc/8.0.1/bin/ghc Example.hs -fprint-explicit-kinds [1 of 1] Compiling Example ( Example.hs, Example.o ) Example.hs:9:57: error: • No instance for (Category * c) arising from the 'deriving' clause of a data type declaration Possible fix: use a standalone 'deriving instance' declaration, so you can specify the instance context yourself • When deriving the instance for (Category * (T c)) }}} I know what is happening. Unlike in #8865, where we were deriving `Category` for a concrete type like `Either`, in the above example we are attempting to derive an instance of the form: {{{#!hs instance Category * c => Category (T * c) where ... }}} (using `-fprint-explicit-kinds` syntax). But `validDerivPred` is checking if `sizePred (Category * c)` equals the number of free type variables in `Category * c`. But note that `sizePred` counts both type variables //and// type constructors, and `*` is a type constructor! So `validDerivPred` erroneously rejects the above instance. To fix this behavior, I think we just need to change `validDerivPred` to only consider the //visible// arguments of `Category` (i.e., only `c`). This should be a pretty easy fix - patch incoming. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11833> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #11484: Type synonym using -XTypeInType can't be spliced with TH
by GHC 25 Aug '16

25 Aug '16

#11484: Type synonym using -XTypeInType can't be spliced with TH -------------------------------------+------------------------------------- Reporter: RyanGlScott | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Template | Version: 8.0.1-rc1 Haskell | Keywords: TypeInType | 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: -------------------------------------+------------------------------------- The following code compiles: {{{#!hs {-# LANGUAGE TemplateHaskell #-} {-# LANGUAGE TypeInType #-} {-# OPTIONS_GHC -ddump-splices #-} module Foo where import Data.Kind type TySyn (k :: *) (a :: k) = () -- $([d| type TySyn2 (k :: *) (a :: k) = () |]) }}} But uncomment the last line, and it doesn't compile: {{{ $ /opt/ghc/head/bin/ghc Foo.hs [1 of 1] Compiling Foo ( Foo.hs, Foo.o ) Foo.hs:10:3-43: Splicing declarations [d| type TySyn2_aBH (k_aBI :: *) (a_aBJ :: k_aBI) = () |] ======> type TySyn2_a4BF (k_a4BG :: Type) (a_a4BH :: k_a4BG) = () Foo.hs:10:3: error: • Couldn't match expected kind ‘GHC.Prim.Any’ with actual kind ‘k1’ • In the type declaration for ‘TySyn2’ Foo.hs:10:3: error: • Kind variable ‘k_a4BG’ is implicitly bound in datatype ‘TySyn2’, but does not appear as the kind of any of its type variables. Perhaps you meant to bind it (with TypeInType) explicitly somewhere? Type variables with inferred kinds: k_a4BG (a_a4BH :: GHC.Prim.Any) • In the type declaration for ‘TySyn2’ }}} There are two issues here: 1. The error message claims that `TySyn2` is a datatype when it is actually a type synonym. This should be easy enough to fix; just change [http://git.haskell.org/ghc.git/blob/89bdac7635e6ed08927d760aa885d3e7ef3edb8… the code] that throws the error message to invoke [http://git.haskell.org/ghc.git/blob/89bdac7635e6ed08927d760aa885d3e7ef3edb8… tyConFlavour]. 2. For some reason, the type variable `a` in `TySyn2` fails to kind-check. Somehow, an `Any` got in there, but I'm not sure where it snuck in. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11484> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 10

[GHC] #11976: Error message mentions impredicative polymorphism
by GHC 25 Aug '16

25 Aug '16

#11976: Error message mentions impredicative polymorphism -------------------------------------+------------------------------------- Reporter: Iceland_jack | 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: -------------------------------------+------------------------------------- When we apply `Lens` to too many arguments in the `:kind` command we get a sensible error message {{{ ghci> type Lens s a = forall f. Functor f => (a -> f a) -> (s -> f s) ghci> :kind Lens _ _ Lens _ _ :: * ghci> :kind Lens _ _ _ <interactive>:1:1: error: • Expecting one fewer argument to ‘Lens t0 t1’ Expected kind ‘k0 -> k1’, but ‘Lens t0 t1’ has kind ‘*’ • In the type ‘Lens _ _ _’ }}} but use the same in a type signature or annotation and it complains about impredicative polymorphism {{{ ghci> :t undefined :: Lens _ _ _ <interactive>:1:1: error: • Illegal polymorphic type: Lens _ _1 GHC doesn't yet support impredicative polymorphism • In the expression: undefined :: Lens _ _ _ ghci> a :: Lens _ _ _; a = undefined <interactive>:6:1: error: Illegal polymorphic type: Lens _ _1 GHC doesn't yet support impredicative polymorphism }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11976> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #11941: stage restriction mentioned when an identifier is out of scope
by GHC 25 Aug '16

25 Aug '16

#11941: stage restriction mentioned when an identifier is out of scope -------------------------------------+------------------------------------- Reporter: aavogt | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1-rc2 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 TemplateHaskell #-} import Data.Monoid const (return []) $ mempty { getFrst = Just () } }}} {{{ typeErrBug.hs:5:21: error: • GHC stage restriction: ‘getFrst’ is used in a top-level splice, quasi-quote, or annotation, and must be imported, not defined locally • In the second argument of ‘($)’, namely ‘mempty {getFrst = Just ()}’ In the expression: const (return []) $ mempty {getFrst = Just ()} typeErrBug.hs:5:30: error: Not in scope: ‘getFrst’ Perhaps you meant one of these: ‘getFirst’ (imported from Data.Monoid), ‘getLast’ (imported from Data.Monoid) }}} The stage restriction error shouldn't be mentioned: `getFrst` isn't defined at all while the "not defined locally" phrase says it's defined (in the wrong spot). -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11941> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3