March 2016 - ghc-tickets

[GHC] #9539: TQueue can lead to thread starvation
by GHC 12 Jun '18

12 Jun '18

#9539: TQueue can lead to thread starvation -------------------------------------+------------------------------------- Reporter: jwlato | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: libraries (other) | Version: 7.8.2 Keywords: stm | Operating System: Architecture: Unknown/Multiple | Unknown/Multiple Difficulty: Unknown | Type of failure: Blocked By: | None/Unknown Related Tickets: | Test Case: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- For background, please see [http://stackoverflow.com/questions/25536604 /huge-memory-consumption-for-simple-multithreaded- haskell/25560047#25560047 this SO question] When using TQueue, a sufficiently fast/persistent writer can result in the reader never getting scheduled, causing the queue to continually grow and lost concurrency. I believe the issue is caused by the definition of readTQueue: {{{ readTQueue :: TQueue a -> STM a readTQueue (TQueue read write) = do xs <- readTVar read case xs of (x:xs') -> do writeTVar read xs' return x [] -> do ys <- readTVar write case ys of [] -> retry _ -> case reverse ys of [] -> error "readTQueue" (z:zs) -> do writeTVar write [] writeTVar read zs return z }}} Note that the last `case` needs to traverse the write-end of the queue within the STM transaction. If the list is sufficiently large, a writer can commit a new transaction much more quickly, invalidating the read transaction. If threads continue to write to the queue, the reader will never get an opportunity to commit (and the list will grow, exacerbating the problem). This alternative definition seems to fix the problem, but I don't know if there are other drawbacks to using it. {{{ readTQueue' :: TQueue a -> STM a readTQueue' (TQueue read write) = do xs <- readTVar read case xs of (x:xs') -> do writeTVar read xs' return x [] -> do ys <- readTVar write case ys of [] -> retry _ -> do writeTVar write [] let (z:zs) = reverse ys writeTVar read zs return z }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9539> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 23

[GHC] #11376: Inconsistent specified type variables among functions and datatypes/classes when using -XTypeApplications
by GHC 08 Jun '18

08 Jun '18

#11376: Inconsistent specified type variables among functions and datatypes/classes when using -XTypeApplications -------------------------------------+------------------------------------- Reporter: RyanGlScott | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.1 (Type checker) | Keywords: TypeInType | Operating System: Unknown/Multiple Architecture: | Type of failure: Other Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Originally reported [https://mail.haskell.org/pipermail/ghc- devs/2016-January/010915.html here]. When applying types via `-XTypeApplications`, the type/kind that gets instantiated seems to be different depending on whether you're using functions, datatypes, or typeclasses. Here's an example contrasting functions and datatypes: {{{ $ /opt/ghc/head/bin/ghci GHCi, version 8.1.20160106: http://www.haskell.org/ghc/ :? for help λ> :set -fprint-explicit-kinds -XTypeApplications -XTypeInType λ> data Prox a = Prox λ> prox :: Prox a; prox = Prox λ> :t prox prox :: forall k (a :: k). Prox k a λ> :t prox @Int prox @Int :: Prox * Int λ> :t Prox Prox :: forall k (a :: k). Prox k a λ> :t Prox @Int Prox @Int :: Prox Int a }}} This is the core of the problem: with a function, `Int` seems to be applied to type variable `a`, whereas with data types, `Int` appears to be applied to the kind variable `k`. This confuses me, since `k` wasn't specified anywhere in the definition of `Prox`. Andres Löh also [https://mail.haskell.org/pipermail/ghc- devs/2016-January/010916.html observed] a similar discrepancy between functions and typeclasses: {{{ λ> :set -XScopedTypeVariables λ> class C a where c :: () λ> d :: forall a. C a => (); d = c @_ @a λ> :t d d :: forall k (a :: k). C k a => () λ> :t d @Int d @Int :: C * Int => () λ> :t c c :: forall k (a :: k). C k a => () λ> :t c @Int c @Int :: C Int a => () λ> :t c @_ @Int c @_ @Int :: C * Int => () }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11376> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 37

[GHC] #8516: Add (->) representation and the Invariant class to GHC.Generics
by GHC 04 Jun '18

04 Jun '18

#8516: Add (->) representation and the Invariant class to GHC.Generics ----------------------------------------------+---------------------------- Reporter: nfrisby | Owner: Type: feature request | Status: new Priority: low | Milestone: Component: Compiler (Type checker) | Version: 7.7 Keywords: | Operating System: Architecture: Unknown/Multiple | Unknown/Multiple Difficulty: Moderate (less than a day) | Type of failure: Blocked By: | None/Unknown Related Tickets: | Test Case: | Blocking: ----------------------------------------------+---------------------------- We currently disallow any use of the parameter in the domain of (->). {{{ newtype F a = F ((a -> Int) -> Int) deriving Generic1 <interactive>:4:38: Can't make a derived instance of `Generic1 (F g)': Constructor `F' must use the last type parameter only as the last argument of a data type, newtype, or (->) In the data declaration for `F' }}} !DeriveFunctor succeeds for this F. I'd like to add this representation type to GHC.Generics and !DeriveGeneric. {{{ newtype (f :->: g) a = FArrow1 (f a -> g a) }}} We could then represent the first example above. We could also derive the more interesting Generic1 (F g). {{{ newtype F g a = F (g a -> Int) deriving Generic1 type instance Rep1 (F g) = Rec1 g :->: Rec0 Int instance Generic1 (F g) where to x = F $ unRec0 . unArrow1 x . Rec1 from (F x) = FArrow1 $ Rec0 . x . unRec1 }}} Admittedly, there's not many generic definitions impeded by not having (:->:). Contra- and in-variant types are uncommon. I'm suggesting this feature without strong motivating examples because I think this would streamline the implementation of -XDeriveGenerics in some ways while also making it more general — assuming that we added the Invariant class to base or ghc-prim. {{{ class Invariant t where invmap :: (a -> b) -> (b -> a) -> t a -> t b invmap_covariant :: Functor t => (a -> b) -> (b -> a) -> t a -> t b invmap_covariant f _ = fmap f instance (Invariant f,Invariant g) => Invariant (FArrow f g) where invmap co contra (FArrow h) = FArrow $ invmap co contra . h . invmap contra co }}} (Of course, Invariant should be a super class of Functor. :/ ) Now we can handle quite involved examples: {{{ newtype F g h a = F (g (h a)) deriving Generic1 instance Invariant g => Generic1 (F g h) where to x = invmap unRec1 Rec1 $ unComp1 x from (F x) = Comp1 $ invmap Rec1 unRec1 }}} All of that said, I'm mostly opening this ticket so I can get feedback on difficulties I might not be anticipating and have a place to reference from the compiler source code comments. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8516> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 18

[GHC] #9247: Document -XDatatypeContexts in flag reference
by GHC 03 Jun '18

03 Jun '18

#9247: Document -XDatatypeContexts in flag reference ------------------------------------+------------------------------------- Reporter: goldfire | Owner: goldfire Type: bug | Status: new Priority: low | Milestone: Component: Documentation | Version: 7.8.2 Keywords: | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- It may be a deprecated feature, but it still exists and should be listed. I can do this. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9247> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #11066: Inacessible branch should be warning - otherwise breaks type soundness?
by GHC 03 Jun '18

03 Jun '18

#11066: Inacessible branch should be warning - otherwise breaks type soundness? -------------------------------------+------------------------------------- Reporter: rrnewton | Owner: Type: bug | Status: new Priority: high | Milestone: Component: Compiler | Version: 7.10.2 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: -------------------------------------+------------------------------------- GHC 7.10.2 reports inaccessible branches as a type error, and I think this is really a bug. First, the implicit thinking seems to be "why would a user want to write inaccessible cases, they'd be crazy!". But that only applies to human-written code. Program generators can have a great deal of trouble avoiding this error, unless they replicate the GHC type checker to predict the problem and prune branches. (That's why we are hitting this error and are stuck.) Second, it seems like this is a problem for type soundness. See the attached program where "step1" typechecks but "step2" does not. And yet, the operational semantics should allow step1 to reduce to step2. Indeed, in the "GADTs meet their match" paper it seems like the intent was to warn for these inaccessible cases, not error. For example, the paper contains the language: "If we warn that a right-hand side of a non-redundant clause is inaccessible," -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11066> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 50

[GHC] #9724: reexport IsList class from a trustworthy module
by GHC 01 Jun '18

01 Jun '18

#9724: reexport IsList class from a trustworthy module -------------------------------------+------------------------------------- Reporter: int-e | Owner: ekmett Type: feature request | Status: new Priority: normal | Milestone: Component: Core Libraries | Version: 7.8.3 Keywords: | Operating System: Architecture: Unknown/Multiple | Unknown/Multiple Difficulty: Unknown | Type of failure: Blocked By: | None/Unknown Related Tickets: | Test Case: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- ==== Motivation GHC's Safe Haskell extension relies to a large extent on inferring safety of modules automatically. As a concrete example, [[http://hackage.haskell.org/package/data-default|data-default]] used to be inferred safe. This is no longer the case, because [[http://hackage.haskell.org/package/dlist-0.7.1|dlist-0.7.1]] added an {{{IsList}}} instance for difference lists. The only way to do that currently is to import the relevant parts of {{{GHC.Exts}}}, and that module is not safe. Of course the resulting non-safety of the {{{Data.DList}}} module can be fixed by adding a {{{TrustWorthy}}} pragma to the module and trusting the {{{dlist}}} package. But that is unsatisfactory, because there is nothing unsafe about providing an instance for the {{{IsList}}} class. This issue is bound to pop up in other packages as well. ==== Proposed Solution (this is one of many possibilities) Add a {{{GHC.Exts.Safe}}} module that is marked {{{TrustWorthy}}} and reexports the {{{IsList}}} class, and possibly other things from {{{GHC.Exts}}} that are safe to use. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9724> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #9479: Report required constraints when reporting the type of a hole
by GHC 31 May '18

31 May '18

#9479: Report required constraints when reporting the type of a hole -------------------------------------+------------------------------------- Reporter: dominiquedevriese | Owner: Type: feature request | Status: new Priority: low | Milestone: Component: Compiler (Type | Version: 7.8.3 checker) | Operating System: Keywords: holes | Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: Difficulty: Unknown | None/Unknown Blocked By: | Test Case: Related Tickets: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- For the following code {{{ module Test where test :: String test = show _h }}} GHC currently reports: {{{ Found hole ‘_h’ with type: a0 Where: ‘a0’ is an ambiguous type variable Relevant bindings include test :: String (bound at /tmp/Test.hs:4:1) In the first argument of ‘show’, namely ‘_h’ In the expression: show _h In an equation for ‘test’: test = show _h }}} It correctly does not report the lack of a `Show _a` instance as a separate type error. However, it would be useful if the report containing the type of the hole would also contain the constraints that apply to its type. Something like: {{{ Found hole ‘_h’ with type: a0 Where: ‘a0’ is an ambiguous type variable Applicable constraints: Show a0 Relevant bindings include test :: String (bound at /tmp/Test.hs:4:1) In the first argument of ‘show’, namely ‘_h’ In the expression: show _h In an equation for ‘test’: test = show _h }}} I am explicitly *not* suggesting to report a type like `Show a0 => a0` for the hole, because that might mistakenly suggest that we are looking for a value of type `forall a0. Show a0 => a0`, which we are not. A possible alternative is to use an imaginary exists type like `exists a0. Show a0 => a0` but that's probably just even more confusing. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9479> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 11

[GHC] #9438: panic: loading archives not supported
by GHC 29 May '18

29 May '18

#9438: panic: loading archives not supported ----------------------------------+---------------------------------- Reporter: egl | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.3 Keywords: | Operating System: MacOS X Architecture: x86_64 (amd64) | Type of failure: Other Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Revisions: ----------------------------------+---------------------------------- Running {{{ ghci -L/opt/local/lib -latlas }}} eventually leads to {{{ Loading object (static archive) /opt/local/lib/libatlas.a ... ghc: panic! (the 'impossible' happened) (GHC version 7.8.3 for x86_64-apple-darwin): Loading archives not supported Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug }}} This looks similar to 8164 and 4828, which are marked closed. Mac OSX 10.9.4 Xcode 5.1.1 -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9438> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 12

[GHC] #10843: Allow do blocks without dollar signs as arguments
by GHC 29 May '18

29 May '18

#10843: Allow do blocks without dollar signs as arguments -------------------------------------+------------------------------------- Reporter: agibiansky | Owner: agibiansky Type: feature | Status: new request | Priority: normal | Milestone: Component: Compiler | Version: 7.10.2 (Parser) | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- I would like the following to be valid Haskell code: {{{#!hs main = when True do putStrLn "Hello!" }}} Instead of requiring a dollar sign before the "do". This would parse as {{{#!hs main = when True (do putStrLn "Hello!") }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10843> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 48

[GHC] #9793: Some as-patterns could be accepted in pattern synonyms
by GHC 29 May '18

29 May '18

#9793: Some as-patterns could be accepted in pattern synonyms -------------------------------------+------------------------------------- Reporter: cactus | Owner: cactus Type: feature request | Status: new Priority: normal | Milestone: 7.10.1 Component: Compiler (Type | Version: 7.8.3 checker) | Operating System: Keywords: pattern synonyms | Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: Difficulty: Easy (less than 1 | None/Unknown hour) | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Revisions: -------------------------------------+------------------------------------- Currently all as-patterns are rejected in pattern synonym definitions, to avoid situations like {{{ pattern P x y <- x@(y:_) }}} since there's no valid reason to be able to then write something like {{{ f [True] False = ... }}} this would just lead to confusion. However, I think we could accept as-patterns where the body of the as- pattern doesn't contain any variable bindings that are accessible via the pattern synonym. In other words, this should be OK: {{{ pattern P x <- x@(y:_) }}} since it's exactly equivalent to {{{ pattern P x <- x@(_:_) }}} which I don't think is as confusing as the other example. I haven't really made up my mind yet if these should be bidirectional; but they certainly could be, by just ignoring the body of the as-pattern in the wrapper; so the following two would be equivalent: {{{ pattern P1 x = [x@(y:_)] pattern P2 x <- [x@(y:_)] where P2 x = [x] }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9793> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 13