November 2016 - ghc-tickets

[GHC] #12876: Appreciate your work!
by GHC 25 Nov '16

25 Nov '16

#12876: Appreciate your work! -------------------------------------+------------------------------------- Reporter: deech | 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: -------------------------------------+------------------------------------- Not a bug. Happy Thanksgiving and thanks for all your efforts! -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12876> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

[GHC] #12867: Internal error while typechecking invalid program
by GHC 25 Nov '16

25 Nov '16

#12867: Internal error while typechecking invalid program -------------------------------------+------------------------------------- Reporter: bgamari | Owner: Type: bug | Status: new Priority: normal | Milestone: 8.2.1 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: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- While trying to reduce #12866 I encountered this program, {{{#!hs {-# LANGUAGE TypeFamilies #-} {-# LANGUAGE ConstraintKinds #-} {-# LANGUAGE FlexibleContexts #-} module T12866 where type Test2 a = (Eq (TestM a)) class Test a where type TestM :: * }}} which gives rise to this internal GHC error during typechecking, {{{ $ ghc Hi2.hs Hi2.hs:7:21: error: • GHC internal error: ‘TestM’ is not in scope during type checking, but it passed the renamer tcl_env of environment: [ap0 :-> Type variable ‘a’ = a] • In the first argument of ‘Eq’, namely ‘TestM a’ In the type ‘Eq (TestM a)’ In the type declaration for ‘Test2’ Hi2.hs:9:1: error: • The associated type ‘TestM’ mentions none of the type or kind variables of the class ‘Test a’ • In the class declaration for ‘Test’ }}} Clearly the program is bogus, but the first error message is quite suspicious. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12867> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #12845: -XPartialTypeSignatures results in unbound variables
by GHC 25 Nov '16

25 Nov '16

#12845: -XPartialTypeSignatures results in unbound variables -------------------------------------+------------------------------------- Reporter: crockeea | 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: -------------------------------------+------------------------------------- The following code fails to compile, but I think it should. {{{ {-# LANGUAGE DataKinds #-} {-# LANGUAGE PartialTypeSignatures #-} {-# LANGUAGE ScopedTypeVariables #-} {-# LANGUAGE TypeFamilies #-} {-# LANGUAGE TypeOperators #-} import Data.Proxy data Foo (m :: Bool) type family Head (xs :: [(Bool, Bool)]) where Head (x ': xs) = x type family Bar (x :: Bool) (y :: Bool) :: Bool -- to trigger the bug, r and r' cannot *both* appear on the RHS broken :: forall r r' rngs . ('(r,r') ~ Head rngs, Bar r r' ~ 'True, _) => Foo r -> Proxy rngs -> () broken x _ = let y = requireBar x :: Foo r' in () requireBar :: (Bar m m' ~ 'True) => Foo m -> Foo m' requireBar = undefined }}} The error is: {{{ > ghci UnboundVar.hs GHCi, version 8.0.1: http://www.haskell.org/ghc/ :? for help [1 of 1] Compiling Main ( UnboundVar.hs, interpreted ) UnboundVar.hs:18:1: error: • Could not deduce: Bar r r'0 ~ 'True from the context: ('(r, r') ~ Head rngs, Bar r r' ~ 'True, Bar r r' ~ 'True) bound by the inferred type for ‘broken’: ('(r, r') ~ Head rngs, Bar r r' ~ 'True, Bar r r' ~ 'True) => Foo r -> Proxy rngs -> () at UnboundVar.hs:18:1-49 The type variable ‘r'0’ is ambiguous • When checking that the inferred type broken :: forall (r :: Bool) (r' :: Bool) (rngs :: [(Bool, Bool)]). Bar r r' ~ 'True => Foo r -> Proxy rngs -> () is as general as its (partial) signature broken :: forall (r :: Bool) (r' :: Bool) (rngs :: [(Bool, Bool)]). ('(r, r') ~ Head rngs, Bar r r' ~ 'True, Bar r r' ~ 'True) => Foo r -> Proxy rngs -> () }}} If I remove the partial type signature (i.e., remove the `_` constraint from `broken`), then GHC happily accepts the code. I'm not sure why adding a `_` introduces ambiguity; it seems GHC is forgetting that `r'` is bound by the `forall` (and that `-XScopedTypeVariables` is enabled). -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12845> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #12826: TyVar ASSERT failure in type family checking: T12041
by GHC 25 Nov '16

25 Nov '16

#12826: TyVar ASSERT failure in type family checking: T12041 -------------------------------------+------------------------------------- Reporter: simonpj | 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: -------------------------------------+------------------------------------- In HEAD, if you build the compiler with -DDEBUG, test `indexed- types/should_fail/T12041` fails thus {{{ +ghc: panic! (the 'impossible' happened) + (GHC version 8.1.20161109 for x86_64-unknown-linux): + ASSERT failed! + i_axb + Call stack: + CallStack (from HasCallStack): + prettyCurrentCallStack, called at compiler/utils/Outputable.hs:<line>:<column> in <package-id>:Outputable + callStackDoc, called at compiler/utils/Outputable.hs:<line>:<column> in <package-id>:Outputable + assertPprPanic, called at compiler/typecheck/TcType.hs:<line>:<column> in <package-id>:TcType + Call stack: + CallStack (from HasCallStack): + prettyCurrentCallStack, called at compiler/utils/Outputable.hs:<line>:<column> in <package-id>:Outputable + callStackDoc, called at compiler/utils/Outputable.hs:<line>:<column> in <package-id>:Outputable + pprPanic, called at compiler/utils/Outputable.hs:<line>:<column> in <package-id>:Outputable + assertPprPanic, called at compiler/typecheck/TcType.hs:<line>:<column> in <package-id>:TcType }}} A `TyVar` is showing up where a `TcTyVar` is expected. I'm pretty certain this is harmless in a production compiler, but reflects an infelicity somewhere. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12826> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #12821: Find the incredible vanishing [The improvement story]
by GHC 25 Nov '16

25 Nov '16

#12821: Find the incredible vanishing [The improvement story] -------------------------------------+------------------------------------- Reporter: bgamari | Owner: Type: task | Status: new 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: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Somehow we lost Note [The improvement story] in `TcInteract`. There are references to it in `TcInteract` and `TcRnTypes` but the note itself is nowhere to be found. Someone must find it before it reaches the county border! -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12821> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #12526: regression in type inference with respect to type families
by GHC 25 Nov '16

25 Nov '16

#12526: regression in type inference with respect to type families -------------------------------------+------------------------------------- Reporter: Lemming | Owner: Type: bug | Status: new Priority: normal | Milestone: 8.0.2 Component: Compiler | Version: 8.1 (Type checker) | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: GHC rejects Unknown/Multiple | valid program Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- The following bug is present in ghc-8.0.1.20160725, ghc-8.0.1.20160822 (ghc-8.0.2) and ghc-8.1.20160819 (ghc-head) and was not there in ghc-8.0.1. When compiling the head version of synthesizer-llvm package from http://code.haskell.org/synthesizer/llvm/ I get the following error: {{{ src/Synthesizer/LLVM/Server/CausalPacked/Instrument.hs:296:23: error: • Couldn't match type ‘Synthesizer.Causal.Class.ProcessOf t0’ with ‘CausalP.T (SampleRate Real, ())’ Expected type: CausalP.T (SampleRate Real, ()) (Serial.T (LLVM.Value (LLVM.Vector (TypeNum.Pos Type.Data.Num.Decimal.Digit.Dec4 TypeNum.EndDesc) Float)), (Serial.Value (TypeNum.Pos Type.Data.Num.Decimal.Digit.Dec4 TypeNum.EndDesc) Real, VectorValue)) (Serial.T (LLVM.Value (LLVM.Vector (TypeNum.Pos Type.Data.Num.Decimal.Digit.Dec4 TypeNum.EndDesc) Float))) Actual type: Synthesizer.Causal.Class.ProcessOf t0 (Serial.T (LLVM.Value (LLVM.Vector (TypeNum.Pos Type.Data.Num.Decimal.Digit.Dec4 TypeNum.EndDesc) Float)), (Serial.Value (TypeNum.Pos Type.Data.Num.Decimal.Digit.Dec4 TypeNum.EndDesc) Float, Serial.Value (TypeNum.Pos Type.Data.Num.Decimal.Digit.Dec4 TypeNum.EndDesc) Float)) (Serial.T (LLVM.Value (LLVM.Vector (TypeNum.Pos Type.Data.Num.Decimal.Digit.Dec4 TypeNum.EndDesc) Float))) The type variable ‘t0’ is ambiguous • In the first argument of ‘($&)’, namely ‘osci’ In the second argument of ‘liftA2’, namely ‘(osci $& shapeCtrl &|& (expo &|& fmap Stereo.left freqs))’ In the expression: liftA2 Stereo.cons (osci $& shapeCtrl &|& (expo &|& fmap Stereo.left freqs)) (osci $& shapeCtrl &|& (negate expo &|& fmap Stereo.right freqs)) • Relevant bindings include osci :: Synthesizer.Causal.Class.ProcessOf t0 (Serial.T (LLVM.Value (LLVM.Vector (TypeNum.Pos Type.Data.Num.Decimal.Digit.Dec4 TypeNum.EndDesc) Float)), (Serial.Value (TypeNum.Pos Type.Data.Num.Decimal.Digit.Dec4 TypeNum.EndDesc) Float, Serial.Value (TypeNum.Pos Type.Data.Num.Decimal.Digit.Dec4 TypeNum.EndDesc) Float)) (Serial.T (LLVM.Value (LLVM.Vector (TypeNum.Pos Type.Data.Num.Decimal.Digit.Dec4 TypeNum.EndDesc) Float))) (bound at src/Synthesizer/LLVM/Server/CausalPacked/Instrument.hs:294:19) }}} The type for `t0` should be known and then the type checker should be able to check that `ProcessOf t0` equals `CausalP.T`. It did so in earlier versions but now fails on that part. I still have to reduce that example, but for now I want to alert you that there seems to be a regression in the type inference. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12526> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 16

[GHC] #12538: Incorrect uses of overlapping instances and data families sends GHC into loop
by GHC 25 Nov '16

25 Nov '16

#12538: Incorrect uses of overlapping instances and data families sends GHC into loop -------------------------------------+------------------------------------- Reporter: pkmx | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.1 (Type checker) | 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: -------------------------------------+------------------------------------- Sorry for the lack of descriptive title, as I can't nail down the source of this specific bug. This is the minimal example to trigger the loop: {{{#!hs {-# LANGUAGE CPP #-} {-# LANGUAGE FlexibleInstances #-} {-# LANGUAGE FunctionalDependencies #-} {-# LANGUAGE MultiParamTypeClasses #-} {-# LANGUAGE TypeApplications #-} {-# LANGUAGE TypeFamilies #-} {-# LANGUAGE UndecidableInstances #-} module Main where import GHC.TypeLits import GHC.Types data Tagged t a = Tagged a type family Tag a where Tag (Tagged t a) = Tagged t a Tag a = Tagged Int a class (r ~ Tag a) => TagImpl a r | a -> r where tag :: a -> r instance {-# OVERLAPPING #-} (r ~ Tag (Tagged t a)) => TagImpl (Tagged t a) r where tag = id #ifdef WRONG instance {-# OVERLAPPING #-} (r ~ Tagged t a, r ~ Tag a) => TagImpl a r where #else instance {-# OVERLAPPING #-} (r ~ Tagged Int a, r ~ Tag a) => TagImpl a r where #endif tag = Tagged @Int data family DF x data instance DF (Tagged t a) = DF (Tagged t a) class ToDF a b | a -> b where df :: a -> b #ifdef WRONG instance (TagImpl a a', b ~ DF a') => ToDF a b where #else instance (TagImpl a (Tagged t a'), b ~ DF (Tagged t a')) => ToDF a b where #endif df = DF . tag main :: IO () main = pure () }}} When compiled with `-DWRONG`, it causes GHC (both 8.0.1 and HEAD@20160823) to loop: {{{ $ ghc --version && ghc -fno-code Main.hs -DWRONG The Glorious Glasgow Haskell Compilation System, version 8.1.20160823 [1 of 1] Compiling Main ( Main.hs, nothing ) (loops indefinitely...) }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12538> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #12875: GHC fails to link all StaticPointers-defining modules of a library in an executable
by GHC 24 Nov '16

24 Nov '16

#12875: GHC fails to link all StaticPointers-defining modules of a library in an executable -------------------------------------+------------------------------------- Reporter: bgamari | Owner: Type: bug | Status: new Priority: normal | Milestone: 8.2.1 Component: Compiler | Version: 8.0.1 Keywords: | Operating System: Unknown/Multiple StaticPointers | Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Consider that you have a package called `lib` which exposes these modules, {{{#!hs module ALib.Types where data AThing = AThing String deriving (Show) {-# LANGUAGE StaticPointers #-} module ALib.Things where import GHC.StaticPtr import ALib.Types thing1 :: StaticPtr AThing thing1 = static (AThing "hello") }}} Now consider that you have a server which reads a `StaticKey` of `StaticPtr AThing` and shows it, {{{#!hs import ALib.Types main :: IO () main = do key <- readFingerprint <$> getContents :: IO StaticKey Just thing <- unsafeLookupStaticPtr key :: IO (Maybe (StaticPtr AThing)) print $ deRefStaticPtr thing }}} Naturally this executable will link against `lib`. However, this executable as-written will fail if given the key of `ALib.Things.thing1`. Fixing this requires that the executable explicitly import and use a definition from `ALib.Things`. The problem appears to be that the linker elides the `ALib.Things` object from the final executable unless it refers to a symbol. Note that things also work fine if the server executable is dynamically linked. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12875> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #12864: Produce type errors after looking at whole applications
by GHC 23 Nov '16

23 Nov '16

#12864: Produce type errors after looking at whole applications -------------------------------------+------------------------------------- Reporter: nomeata | Owner: 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: Poor/confusing Unknown/Multiple | error message Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Currently, the type error messages we produce in case of function applications are rather dumb, in particular if arguments were ommited, added, swapped or not parenthized correctly. (Examples in the first comment below.) In many cases the error messages would be better if it would at one application if a function at once, and, in general, present the user with the (inferred) type of the function as well as the (expected) type due to the argument. This way, only one type error will be reported per function application, and comparing these two reported types will allow the user to spot the problem more easily. (With „one application“ I mean `e1 e2 e3 … en` where `e1` is not of that form.) Furthermore, once we have these two types in our hands, we can look for common patterns, and give even more helpful error messages, such as suggesting to swap two arguments. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12864> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

Re: [GHC] #3384: Add HsSyn prettyprinter tests
by GHC 23 Nov '16

23 Nov '16

#3384: Add HsSyn prettyprinter tests -------------------------------------+------------------------------------- Reporter: igloo | Owner: Type: task | Status: new Priority: normal | Milestone: 8.2.1 Component: Test Suite | Version: Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Phab:D2752 Wiki Page: | -------------------------------------+------------------------------------- Changes (by alanz): * differential: => Phab:D2752 -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/3384#comment:14> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0