- ghc-tickets - Haskell.org

[GHC] #11330: Test `dynamic-paper` fails with core lint error (hpc) and "Simplifier ticks exhausted" (optasm)
by GHC 07 Mar '18

07 Mar '18

#11330: Test `dynamic-paper` fails with core lint error (hpc) and "Simplifier ticks exhausted" (optasm) -------------------------------------+------------------------------------- Reporter: thomie | Owner: Type: bug | Status: new Priority: normal | Milestone: 8.0.1 Component: Compiler | Version: 7.11 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: -------------------------------------+------------------------------------- {{{ $ ghc-head --info | grep Project [("Project name","The Glorious Glasgow Haskell Compilation System") ,("Project version","8.1.20151231") ,("Project Git commit id","8afeaad919dc67643b4eff14efafb48b59039b2b") }}} {{{ $ make slowtest TEST=dynamic-paper VERBOSE=2 TEST_HC=ghc-head ... =====> dynamic-paper(normal) 1 of 1 [0, 0, 0] =====> dynamic-paper(hpc) 1 of 1 [0, 0, 0] ... Compile failed (status 256) errors were: *** Core Lint errors : in result of Float out(FOS {Lam = Just 0, Consts = True, OverSatApps = True}) *** <no location info>: warning: In a case alternative: (TypeRepX k_s7Ec :: *, a_s7Ed :: k_a5px, tr2_s7Ee :: TypeRep a_s7Ed) @ k_a5px is out of scope *** Offending Program *** ... <no location info>: error: Compilation had errors *** unexpected failure for dynamic-paper(hpc) =====> dynamic-paper(optasm) 1 of 1 [0, 1, 0] Compile failed (status 256) errors were: ghc: panic! (the 'impossible' happened) (GHC version 8.1.20151231 for x86_64-unknown-linux): Simplifier ticks exhausted When trying UnfoldingDone delta1 To increase the limit, use -fsimpl-tick-factor=N (default 100) If you need to do this, let GHC HQ know, and what factor you needed To see detailed counts use -ddump-simpl-stats Total ticks: 203160 Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug *** unexpected failure for dynamic-paper(optasm) Unexpected results from: TEST="dynamic-paper" }}} This test was added in commit 52da6bdc17bb491d6d2f462b3680eb44b9be92e5: {{{ Author: Richard Eisenberg <eir(a)cis.upenn.edu> Date: Sat Dec 26 09:11:33 2015 -0500 Have mkCastTy look more closely for reflexivity. This may have performance implications. }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11330> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 10

[GHC] #12506: Compile time regression in GHC 8.
by GHC 07 Mar '18

07 Mar '18

#12506: Compile time regression in GHC 8. -------------------------------------+------------------------------------- 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: -------------------------------------+------------------------------------- This bug is entirely reproducible but unfortunately takes some setup. I don't have a minimal example. The regression is happening in FLTKHS (https://github.com/deech/fltkhs) the Haskell bindings to the FLTK(http://fltk.org) toolkit, and in particular to this GUI (https://github.com/deech/fltkhs/blob/master/images /tree-complex-windows.png). In GHC 7.8, it taks 5-9 seconds to compile and link in 7.10.3, that goes up to 12-15 seconds and finally in GHC 8 it shoots up to 45 seconds. While these numbers come from my i7, 16GB RAM Linux laptop, the compile time regression is proportionally reproducible across machines and operating systems. Here are the steps to reproduce it * Install FLTK Please follow the instructions here (https://github.com/deech/fltkhs #linux-and-bsd) but only the first set which about for installing FLTK 1.3. * Clone the repo with the offending application: `git clone https://github.com/deech/fltkhs-fluid-demos` * Build the application and reproduce the problem: - `cabal install fltkhs` - At the root of the repo do `cabal build fltkhs-fluid-tree` -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12506> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 34

Re: [GHC] #7206: Implement cheap build
by GHC 07 Mar '18

07 Mar '18

#7206: Implement cheap build -------------------------------------+------------------------------------- Reporter: simonpj | Owner: simonpj Type: bug | Status: new Priority: normal | Milestone: ⊥ Component: Compiler | Version: 7.4.2 Resolution: | Keywords: FloatOut Operating System: Unknown/Multiple | Architecture: Type of failure: Runtime | Unknown/Multiple performance bug | Test Case: Blocked By: | Blocking: Related Tickets: #8763, #13422 | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by osa1): I confirmed that this fixes #8763. I had not encountered this in practice myself, but I was surprised to see this function used in another bug report: {{{ ireplicateA_ :: Applicative m => Int -> (Int -> m a) -> m () ireplicateA_ cnt0 f = loop cnt0 0 where loop !cnt !n | cnt <= 0 = pure () | otherwise = f n *> (loop (cnt - 1) (n + 1)) }}} when I tried using `forM_ [1 .. n] ...` instead of this function I realized that the program gets much slower. However with `cheapBuild` the generated code is identical. I'll submit a patch with cheap build later today with a perf test. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7206#comment:29> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #7206: Implement cheap build
by GHC 07 Mar '18

07 Mar '18

#7206: Implement cheap build -------------------------------------+------------------------------------- Reporter: simonpj | Owner: simonpj Type: bug | Status: new Priority: normal | Milestone: ⊥ Component: Compiler | Version: 7.4.2 Resolution: | Keywords: FloatOut Operating System: Unknown/Multiple | Architecture: Type of failure: Runtime | Unknown/Multiple performance bug | Test Case: Blocked By: | Blocking: Related Tickets: #8763, #13422 | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by osa1): > Some examples/tickets (which specifically?) really do improve. I don't know which examples/tickets improve. Perhaps @dfeuer knows? -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7206#comment:28> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #14785: accumArray is too lazy
by GHC 07 Mar '18

07 Mar '18

#14785: accumArray is too lazy -------------------------------------+------------------------------------- Reporter: dfeuer | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: 8.6.1 Component: Core | Version: 8.2.2 Libraries | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Other Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- According to the documentation, {{{#!hs -- If the accumulating function is strict, then 'accumArray' is strict in -- the values, as well as the indices, in the association list. Thus, -- unlike ordinary arrays built with 'array', accumulated arrays should -- not in general be recursive. }}} I believe that strictness is desirable, because it prevents thunks from accumulating when we use the passed function. But the strictness isn't actually there! We never force the result of the function. I believe we should fix the code (and documentation) to maintain the invariant that all elements in the result of `accumArray` are in WHNF. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14785> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #14878: Can't witness transitivity ((.)) of isomorphism of Constraints
by GHC 06 Mar '18

06 Mar '18

#14878: Can't witness transitivity ((.)) of isomorphism of Constraints -------------------------------------+------------------------------------- Reporter: Iceland_jack | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.5 Keywords: | Operating System: Unknown/Multiple QuantifiedConstraints, wipT2893 | Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- This compiles, {{{#!hs {-# Language QuantifiedConstraints, GADTs, ConstraintKinds, TypeOperators, FlexibleInstances, MultiParamTypeClasses, UndecidableInstances #-} data Dict c where Dict :: c => Dict c class (a => b) => Implies a b instance (a => b) => Implies a b type a -:- b = Dict (Implies a b, Implies b a) -- isomorphism of constraints, should be an equivalence relation id_ :: a-:-a id_ = Dict sym_ :: a-:-b -> b-:-a sym_ Dict = Dict -- comp_ :: a-:-b -> b-:-c -> a-:-c -- comp_ Dict Dict = Dict }}} but uncommenting `comp_` and GHC doesn't know how to deduce `b` (the location message is weird) {{{ $ ghci -ignore-dot-ghci hs/206-bug-quantified-constraints.hs GHCi, version 8.5.20180128: http://www.haskell.org/ghc/ :? for help [1 of 1] Compiling Main ( hs/206-bug-quantified- constraints.hs, interpreted ) hs/206-bug-quantified-constraints.hs:1:1: error: Could not deduce: b from the context: (Implies a b, Implies b a) bound by a pattern with constructor: Dict :: forall (c :: Constraint). c => Dict c, in an equation for ‘comp_’ at hs/206-bug-quantified-constraints.hs:18:7-10 or from: (Implies b c, Implies c b) bound by a pattern with constructor: Dict :: forall (c :: Constraint). c => Dict c, in an equation for ‘comp_’ at hs/206-bug-quantified-constraints.hs:18:12-15 or from: c bound by a quantified context at hs/206-bug-quantified-constraints.hs:1:1 | 1 | {-# Language QuantifiedConstraints, GADTs, ConstraintKinds, TypeOperators, FlexibleInstances, MultiParamTypeClasses, UndecidableInstances #-} | ^ hs/206-bug-quantified-constraints.hs:1:1: error: Could not deduce: b from the context: (Implies a b, Implies b a) bound by a pattern with constructor: Dict :: forall (c :: Constraint). c => Dict c, in an equation for ‘comp_’ at hs/206-bug-quantified-constraints.hs:18:7-10 or from: (Implies b c, Implies c b) bound by a pattern with constructor: Dict :: forall (c :: Constraint). c => Dict c, in an equation for ‘comp_’ at hs/206-bug-quantified-constraints.hs:18:12-15 or from: a bound by a quantified context at hs/206-bug-quantified-constraints.hs:1:1 | 1 | {-# Language QuantifiedConstraints, GADTs, ConstraintKinds, TypeOperators, FlexibleInstances, MultiParamTypeClasses, UndecidableInstances #-} | ^ Failed, no modules loaded. Prelude> }}} simplifying and uncurrying we get a more minimal example {{{#!hs {-# Language QuantifiedConstraints, GADTs, ConstraintKinds, TypeOperators, FlexibleInstances, MultiParamTypeClasses, UndecidableInstances #-} data Dict c where Dict :: c => Dict c f :: (a => b, b => a, b => c, c => b, a => c, c) => Dict a f = Dict }}} giving a different and longer error message {{{ GHCi, version 8.5.20180128: http://www.haskell.org/ghc/ :? for help [1 of 1] Compiling Main ( 206-bug-quantified-constraints.hs, interpreted ) 206-bug-quantified-constraints.hs:6:6: error: • Could not deduce: c0 from the context: (a => b, b => a, b => c, c => b, a => c, c) bound by the type signature for: f :: forall (a :: Constraint) (b :: Constraint) (c :: Constraint). (a => b, b => a, b => c, c => b, a => c, c) => Dict a at 206-bug-quantified-constraints.hs:6:6-58 • In the ambiguity check for ‘f’ To defer the ambiguity check to use sites, enable AllowAmbiguousTypes In the type signature: f :: (a => b, b => a, b => c, c => b, a => c, c) => Dict a | 6 | f :: (a => b, b => a, b => c, c => b, a => c, c) => Dict a | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 206-bug-quantified-constraints.hs:6:6: error: • Could not deduce: (b, c0) from the context: (a => b, b => a, b => c, c => b, a => c, c) bound by the type signature for: f :: forall (a :: Constraint) (b :: Constraint) (c :: Constraint). (a => b, b => a, b => c, c => b, a => c, c) => Dict a at 206-bug-quantified-constraints.hs:6:6-58 or from: a bound by a quantified context at 206-bug-quantified-constraints.hs:6:6-58 • In the type signature: f :: (a => b, b => a, b => c, c => b, a => c, c) => Dict a | 6 | f :: (a => b, b => a, b => c, c => b, a => c, c) => Dict a | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 206-bug-quantified-constraints.hs:6:6: error: • Could not deduce: b0 from the context: (a => b, b => a, b => c, c => b, a => c, c) bound by the type signature for: f :: forall (a :: Constraint) (b :: Constraint) (c :: Constraint). (a => b, b => a, b => c, c => b, a => c, c) => Dict a at 206-bug-quantified-constraints.hs:6:6-58 or from: c0 bound by a quantified context at 206-bug-quantified-constraints.hs:6:6-58 • In the ambiguity check for ‘f’ To defer the ambiguity check to use sites, enable AllowAmbiguousTypes In the type signature: f :: (a => b, b => a, b => c, c => b, a => c, c) => Dict a | 6 | f :: (a => b, b => a, b => c, c => b, a => c, c) => Dict a | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 206-bug-quantified-constraints.hs:6:6: error: • Could not deduce: c0 from the context: (a => b, b => a, b => c, c => b, a => c, c) bound by the type signature for: f :: forall (a :: Constraint) (b :: Constraint) (c :: Constraint). (a => b, b => a, b => c, c => b, a => c, c) => Dict a at 206-bug-quantified-constraints.hs:6:6-58 or from: b0 bound by a quantified context at 206-bug-quantified-constraints.hs:6:6-58 • In the ambiguity check for ‘f’ To defer the ambiguity check to use sites, enable AllowAmbiguousTypes In the type signature: f :: (a => b, b => a, b => c, c => b, a => c, c) => Dict a | 6 | f :: (a => b, b => a, b => c, c => b, a => c, c) => Dict a | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 206-bug-quantified-constraints.hs:6:6: error: • Could not deduce: b from the context: (a => b, b => a, b => c, c => b, a => c, c) bound by the type signature for: f :: forall (a :: Constraint) (b :: Constraint) (c :: Constraint). (a => b, b => a, b => c, c => b, a => c, c) => Dict a at 206-bug-quantified-constraints.hs:6:6-58 or from: b0 bound by a quantified context at 206-bug-quantified-constraints.hs:6:6-58 • In the ambiguity check for ‘f’ To defer the ambiguity check to use sites, enable AllowAmbiguousTypes In the type signature: f :: (a => b, b => a, b => c, c => b, a => c, c) => Dict a | 6 | f :: (a => b, b => a, b => c, c => b, a => c, c) => Dict a | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 206-bug-quantified-constraints.hs:6:6: error: • Could not deduce: (b, b0) from the context: (a => b, b => a, b => c, c => b, a => c, c) bound by the type signature for: f :: forall (a :: Constraint) (b :: Constraint) (c :: Constraint). (a => b, b => a, b => c, c => b, a => c, c) => Dict a at 206-bug-quantified-constraints.hs:6:6-58 or from: a bound by a quantified context at 206-bug-quantified-constraints.hs:6:6-58 • In the type signature: f :: (a => b, b => a, b => c, c => b, a => c, c) => Dict a | 6 | f :: (a => b, b => a, b => c, c => b, a => c, c) => Dict a | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Failed, no modules loaded. Prelude> }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14878> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #14105: ApplicativeDo causes GHC panic on irrefutable list pattern match
by GHC 06 Mar '18

06 Mar '18

#14105: ApplicativeDo causes GHC panic on irrefutable list pattern match -------------------------------------+------------------------------------- Reporter: BoteboTsebo | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.2.1 Keywords: ApplicativeDo | Operating System: Unknown/Multiple Architecture: x86_64 | Type of failure: None/Unknown (amd64) | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- So this just happened: {{{#!hs {-# language ApplicativeDo #-} module Main where main :: IO () main = do [_] <- pure [] pure () }}} Compiling (under Stack) with `stack exec -- ghc --make src/Main.hs` yields {{{ [1 of 1] Compiling Main ( src/Main.hs, src/Main.o ) ghc: panic! (the 'impossible' happened) (GHC version 8.2.1 for x86_64-unknown-linux): isStrictPattern Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug }}} This does not happen on GHC-8.0.2. It does not happen when `ApplicativeDo` is not activated. It also does not happen when the pattern match line is changed to {{{#!hs [] <- pure [] }}} For completeness, the compiler used by `stack` is that which comes with `nightly-2017-08-10`. I am on Linux x64 (Ubuntu). -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14105> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 11

[GHC] #14893: GHC hangs while producing error message for incorrect TypeFamilies application
by GHC 05 Mar '18

05 Mar '18

#14893: GHC hangs while producing error message for incorrect TypeFamilies application -------------------------------------+------------------------------------- Reporter: coopercm | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.2.2 Keywords: | Operating System: Linux Architecture: | Type of failure: Compile-time Unknown/Multiple | crash or panic Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Here's a minimally reproducible example: {{{#!hs {-# LANGUAGE TypeFamilies #-} {-# LANGUAGE MultiParamTypeClasses #-} module Main where newtype Token f a = Token { unToken :: f a } class Class x (f :: * -> *) where type Meta x f :: * type Body x f :: * type Body x f = Token f Meta main :: IO () main = return () }}} This hangs like so {{{ [2 of 2] Compiling Main ( app/Main.hs, .stack- work/dist/x86_64-linux/Cabal-2.0.1.0/build/foo-exe/foo-exe-tmp/Main.o ) }}} I tested on 8.2.2 and 8.0.2 and got the same behavior. The closest existing bug report I could find was https://ghc.haskell.org/trac/ghc/ticket/12386. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14893> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #14892: Field imposters with DuplicateRecordFields
by GHC 05 Mar '18

05 Mar '18

#14892: Field imposters with DuplicateRecordFields -------------------------------------+------------------------------------- Reporter: philderbeast | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.2.2 Keywords: | Operating System: MacOS X Architecture: x86_64 | Type of failure: GHC accepts (amd64) | invalid program Test Case: | Blocked By: Blocking: | Related Tickets: 13644 Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- There's a [[https://github.com/BlockScope/ghc-panic-12158|test case]] for this ghc panic. If I use `NamedFieldPuns` I wish this could more delicately pick the right field. If my record use is qualified then please restrict the set of candidate field names to use punned. {{{ {-# LANGUAGE DuplicateRecordFields #-} module Geodesy (X(..), Y(..)) where data X a = X {x :: a} data Y a = Y {x :: a} }}} {{{ {-# LANGUAGE NamedFieldPuns #-} module GhcPanic12158 where import qualified Geodesy as G (X(..)) import Geodesy (Y(..)) update :: G.X a -> G.X a update G.X{x} = G.X{x = x} }}} {{{ > stack build ghc-panic-translateConPatVec-lookup-0.1.0: build (lib) Preprocessing library for ghc-panic-translateConPatVec-lookup-0.1.0.. Building library for ghc-panic-translateConPatVec-lookup-0.1.0.. [1 of 3] Compiling Geodesy [2 of 3] Compiling GhcPanic12158 ghc: panic! (the 'impossible' happened) (GHC version 8.2.2 for x86_64-apple-darwin): translateConPatVec: lookup Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug }}} There are some fixes; 1. Add DuplicateRecordFields. {{{ {-# LANGUAGE NamedFieldPuns #-} {-# LANGUAGE DuplicateRecordFields #-} module GhcPanic12158 where import qualified Geodesy as G (X(..)) import Geodesy (Y(..)) update :: G.X a -> G.X a update G.X{x} = G.X{x = x} }}} 2. Use qualified field names. {{{ {-# LANGUAGE NamedFieldPuns #-} module GhcPanic12158 where import qualified Geodesy as G (X(..)) import Geodesy (Y(..)) update :: G.X a -> G.X a update G.X{G.x} = G.X{G.x = x} }}} Interestingly, if I don't import the record with the clashing field name then GHC complains. {{{ {-# LANGUAGE NamedFieldPuns #-} module GhcPanic12158 where import qualified Geodesy as G (X(..)) update :: G.X a -> G.X a update G.X{x} = G.X{x = x} }}} {{{ > stack build ghc-panic-translateConPatVec-lookup-0.1.0: unregistering ghc-panic-translateConPatVec-lookup-0.1.0: build (lib) Preprocessing library for ghc-panic-translateConPatVec-lookup-0.1.0.. Building library for ghc-panic-translateConPatVec-lookup-0.1.0.. [3 of 3] Compiling GhcPanic12158 /ghc-panic-12158/earth/library/GhcPanic12158.hs:8:12: error: Not in scope: ‘x’ | 8 | update G.X{x} = G.X{x = x} | ^ /ghc-panic-12158/earth/library/GhcPanic12158.hs:8:21: error: Not in scope: ‘x’ | 8 | update G.X{x} = G.X{x = x} }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14892> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #14840: QuantifiedConstraints: Can't define class alias
by GHC 05 Mar '18

05 Mar '18

#14840: QuantifiedConstraints: Can't define class alias -------------------------------------+------------------------------------- Reporter: Iceland_jack | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.5 Keywords: | Operating System: Unknown/Multiple QuantifiedConstraints wipT2893 | Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- {{{#!hs {-# Language TypeInType, MultiParamTypeClasses, TypeFamilies, FlexibleContexts, FunctionalDependencies, AllowAmbiguousTypes, QuantifiedConstraints, GADTs, ConstraintKinds, KindSignatures, RankNTypes, FlexibleInstances, UndecidableInstances, TypeOperators #-} import Data.Kind class Reifies s a | s -> a where reflect :: proxy s -> a newtype Lift :: forall k. (Type -> Constraint) -> Type -> k -> Type where Lift :: a -> Lift cls a s data family Def :: (k -> Constraint) -> (k -> Type) class (forall s a. Reifies s (Def cls a) => cls (Lift cls a s)) => ReifiableConstraint cls instance (forall s a. Reifies s (Def cls a) => cls (Lift cls a s)) => ReifiableConstraint cls }}} gives an error {{{ $ ghci -ignore-dot-ghci Bug.hs GHCi, version 8.5.20180128: http://www.haskell.org/ghc/ :? for help [1 of 1] Compiling Main ( /tmp/Bug.hs, interpreted ) /tmp/Bug.hs:14:10: error: • Could not deduce: cls (Lift cls a s) arising from the superclasses of an instance declaration from the context: forall (s :: k) a. Reifies s (Def cls a) => cls (Lift cls a s) bound by the instance declaration at /tmp/Bug.hs:14:10-93 or from: Reifies s (Def cls a) bound by a quantified context at /tmp/Bug.hs:1:1 • In the instance declaration for ‘ReifiableConstraint cls’ | 14 | instance (forall s a. Reifies s (Def cls a) => cls (Lift cls a s)) => ReifiableConstraint cls | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Failed, no modules loaded. Prelude> }}} This is a rare occasion that the class alias trick fails for me so ~+~yay~+~. But is it intended? -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14840> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7