June 2018 - ghc-tickets

[GHC] #14978: GADTs don't seem to unpack properly
by GHC 18 Jun '18

18 Jun '18

#14978: GADTs don't seem to unpack properly -------------------------------------+------------------------------------- Reporter: dfeuer | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: 8.6.1 Component: Compiler | Version: 8.2.2 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Runtime Unknown/Multiple | performance bug Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Consider this simple code: {{{#!hs type family Foo a b data Goof a b where Goof :: !(Foo a b :~: Int) -> Goof a b }}} `(:~:)` has just one constructor (`Refl`) with no arguments, so I'd expect it to unpack to a void representation: {{{#!hs GoofConstr :: Foo a b ~N# Int => Goof a b WGoof :: Foo a b :~: Int -> Goof a b WGoof eq = case eq of Refl -> GoofConstr }}} But it doesn't seem to unpack at all. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14978> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #14904: Compiler panic (piResultTy)
by GHC 18 Jun '18

18 Jun '18

#14904: Compiler panic (piResultTy) -------------------------------------+------------------------------------- Reporter: kcsongor | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.2.2 (Type checker) | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: 14873 Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Type-checking the following type family {{{#!hs type family F (f :: forall a. g a) :: Type where F (f :: forall a. g a) = Int }}} panics with the message: {{{#!txt ghc: panic! (the 'impossible' happened) (GHC version 8.4.0.20180118 for x86_64-apple-darwin): piResultTy k_aVM[tau:1] a_aVF[sk:1] Call stack: CallStack (from HasCallStack): callStackDoc, called at compiler/utils/Outputable.hs:1150:37 in ghc:Outputable pprPanic, called at compiler/types/Type.hs:950:35 in ghc:Type }}} The panic happens with HEAD, 8.4 and 8.2. 8.0 rejects the program with an error message, but even it panics on the following version: {{{#!hs type family F f :: Type where F ((f :: forall a. g a) :: forall a. g a) = Int }}} #14873 seemed somewhat related, so I tried with the patch implemented in 3d252037234ce48f9bdada7d5c9b1d8eba470829, but that fails with the same panic too. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14904> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 11

Re: [GHC] #2168: ghci should show haddock comments for identifier
by GHC 18 Jun '18

18 Jun '18

#2168: ghci should show haddock comments for identifier -------------------------------------+------------------------------------- Reporter: j.waldmann | Owner: (none) Type: feature request | Status: closed Priority: normal | Milestone: Component: GHCi | Version: 6.8.2 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: | -------------------------------------+------------------------------------- Changes (by sjakobi): * status: new => closed * resolution: => fixed Comment: The `:doc` command introduced in 85309a3cda367425cca727dfa45e5e6c63b47391 is somewhat limited – it doesn't show docs for function arguments for example. But I to improve it further for GHC-8.8. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/2168#comment:30> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #14845: TypeInType, index by GADT witnessing constraint
by GHC 18 Jun '18

18 Jun '18

#14845: TypeInType, index by GADT witnessing constraint -------------------------------------+------------------------------------- Reporter: Iceland_jack | Owner: (none) Type: feature | Status: new request | Priority: normal | Milestone: Component: Compiler | Version: 8.5 Keywords: TypeInType | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Just wondering if it would ever make sense to allow or use constraints like this {{{#!hs {-# Language PolyKinds, DataKinds, KindSignatures, GADTs, TypeInType, ConstraintKinds #-} import Data.Kind data Struct :: (k -> Constraint) -> (k -> Type) where S :: cls a => Struct cls a data Foo a where F :: Foo (S::Struct Eq a) }}} {{{ $ ghci -ignore-dot-ghci /tmp/test.hs GHCi, version 8.5.20180105: http://www.haskell.org/ghc/ :? for help [1 of 1] Compiling Main ( /tmp/test.hs, interpreted ) /tmp/test.hs:9:13: error: • Illegal constraint in a type: cls0 a0 • In the first argument of ‘Foo’, namely ‘(S :: Struct Eq a)’ In the type ‘Foo (S :: Struct Eq a)’ In the definition of data constructor ‘F’ | 9 | F :: Foo (S::Struct Eq a) | ^ Failed, no modules loaded. Prelude> }}} Please close the ticket if it doesn't -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14845> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 22

[GHC] #14487: Can't Hide Field When DuplicateRecordFields Is Enabled
by GHC 18 Jun '18

18 Jun '18

#14487: Can't Hide Field When DuplicateRecordFields Is Enabled -------------------------------------+------------------------------------- Reporter: iansullivan88 | Owner: (none) Type: bug | Status: new Priority: lowest | Milestone: Component: Compiler | Version: 8.0.2 Keywords: | Operating System: Linux DuplicateRecordFields | Architecture: x86_64 | Type of failure: GHC rejects (amd64) | valid program Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- A.hs fails to compile when DuplicateRecordFields is enabled in B.hs. A.hs {{{#!hs module A where import B hiding (duplicateName) test = X duplicateName duplicateName = 5 }}} B.hs {{{#!hs {-# LANGUAGE DuplicateRecordFields #-} module B where data X = X { duplicateName :: Int } }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14487> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #15264: Warn about implicit kind variables with -Wcompat
by GHC 18 Jun '18

18 Jun '18

#15264: Warn about implicit kind variables with -Wcompat -------------------------------------+------------------------------------- Reporter: int-index | Owner: (none) Type: task | Status: new Priority: high | Milestone: 8.6.1 Component: Compiler | Version: 8.5 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: -------------------------------------+------------------------------------- According to an accepted proposal https://github.com/ghc-proposals/ghc- proposals/blob/master/proposals/0024-no-kind-vars.rst: With `-Wcompat`, warn if a kind variable is brought into scope implicitly in a type with an explicit `forall`. This applies to type signatures and to other contexts that allow a `forall` with the forall-or- nothing rule in effect (for example, class instances). Creating ticket just for the record. Implementation coming soon. Setting priority as high because if the warning doesn't make it into the release, the actual change will have to be delayed by another half a year. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/15264> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #14005: Explore moving from pretty to prettyprinter
by GHC 18 Jun '18

18 Jun '18

#14005: Explore moving from pretty to prettyprinter -------------------------------------+------------------------------------- Reporter: bgamari | Owner: (none) Type: task | Status: new Priority: normal | Milestone: 8.4.1 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: | wiki:PrettyErrors -------------------------------------+------------------------------------- GHC has for a long time used the `pretty` library for pretty-printing. However, for a number of reasons (explained in wiki:PrettyErrors) it leaves some things to be desired. The `prettyprinter` library is a performant, well-supported alternative that looks quite promising. Explore moving to `prettyprinter` -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14005> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 16

[GHC] #15280: StgCmmEnv: variable not found
by GHC 18 Jun '18

18 Jun '18

#15280: StgCmmEnv: variable not found -------------------------------------+------------------------------------- Reporter: wanjach | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.2 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: -------------------------------------+------------------------------------- {{{ stack build warduino-0.1.0.0: build (lib + exe) Preprocessing library warduino-0.1.0.0... Preprocessing executable 'blink' for warduino-0.1.0.0... Preprocessing executable 'lcd' for warduino-0.1.0.0... [1 of 1] Compiling Main ( lcd/Main.hs, .stack-work/dist/x86_64 -linux-nix/Cabal-1.24.2.0/build/lcd/lcd-tmp/Main.o ) /home/user/src/warduino/lcd/Main.hs:31:5: warning: [-Wname-shadowing] This binding for `lcd' shadows the existing binding defined at lcd/Main.hs:30:1 /home/user/src/warduino/lcd/Main.hs:33:9: warning: [-Wname-shadowing] This binding for `print' shadows the existing binding imported from `Prelude' at lcd/Main.hs:2:8-11 (and originally defined in `System.IO') ghc: panic! (the 'impossible' happened) (GHC version 8.0.2 for x86_64-unknown-linux): StgCmmEnv: variable not found lcd_a3jG local binds for: delay_deep'1 $trModule1 $trModule2 delay lcdController $trModule lcd5 lcdController1 lcdController2 lcdController3 lcdController4 lcdController5 lcdController6 lcdController7 lcd6 lcd7 lvl_r9Ee lvl1_r9Ef lvl2_r9Eg lvl3_r9Eh lvl4_r9Ei lvl5_r9Ej lvl6_r9Ek lvl7_r9El delay_deep' ipv1_s9Ev addr#_s9Ew ds4_s9EA sat_s9EB sat_s9EC sat_s9ED Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug -- While building custom Setup.hs for package warduino-0.1.0.0 using: }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/15280> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #9106: GHC Panic related to functional dependencies - kindFunResult
by GHC 18 Jun '18

18 Jun '18

#9106: GHC Panic related to functional dependencies - kindFunResult ----------------------------+--------------------------------------- Reporter: yuriy | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.2 Keywords: | Operating System: Windows Architecture: x86 | Type of failure: Compile-time crash Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ----------------------------+--------------------------------------- Minimal example: {{{ {-# LANGUAGE MultiParamTypeClasses, DataKinds, FunctionalDependencies, TypeOperators, KindSignatures, PolyKinds, FlexibleInstances, FlexibleContexts, UndecidableInstances #-} import GHC.TypeLits data Proxy (a :: k) = Proxy class FunctorN n f a fa | n f a -> fa where fmapn :: Proxy n -> Proxy f -> (a -> a) -> fa -> fa instance FunctorN 0 f a a where fmapn _ _ a = a instance (Functor f, FunctorN (n - 1) f a fa) => FunctorN n f a (f fa) where fmapn _ pf f = fmap (fmapn (Proxy :: Proxy (n-1)) pf f) }}} Crashes with ghc and ghci: {{{ >ghc test [1 of 1] Compiling Main ( test.hs, test.o ) ghc.exe: panic! (the 'impossible' happened) (GHC version 7.8.20140130 for x86_64-unknown-mingw32): kindFunResult k{tv azb} [sk] >ghci test GHCi, version 7.8.20140130: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... done. [1 of 1] Compiling Main ( test.hs, interpreted ) ghc.exe: panic! (the 'impossible' happened) (GHC version 7.8.20140130 for x86_64-unknown-mingw32): kindFunResult k{tv aPm} [sk] }}} Removing the functional dependency makes the code compile. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9106> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 8

[GHC] #15103: Speed optimizations for elimCommonBlocks
by GHC 18 Jun '18

18 Jun '18

#15103: Speed optimizations for elimCommonBlocks -------------------------------------+------------------------------------- Reporter: AndreasK | Owner: (none) Type: task | Status: new Priority: normal | Milestone: 8.6.1 Component: Compiler | Version: 8.5 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): Phab:D4597 | Wiki Page: -------------------------------------+------------------------------------- == Use toBlockList instead of revPostorder. Block elimination works on a given Cmm graph by: * Getting a list of blocks. * Looking for duplicates in these blocks. * Removing all but one instance of duplicates. There are two (reasonable) ways to get the list of blocks. * The fast way: `toBlockList`: This just flattens the underlying map into a list. * The convenient way: `revPostorder`: Start at the entry label, scan for reachable blocks and return only these. This has the advantage of removing all dead code. If there is dead code the later is better. Work done on unreachable blocks is clearly wasted work. However by the point we run the common block elimination pass the input graph already had all dead code removed. This is done during control flow optimization in CmmContFlowOpt which is our first Cmm pass. This means common block elimination is free to use toBlockList because revPostorder would return the same blocks. (Although in a different order). == Change the triemap used for grouping by a label list from `(TM.ListMap UniqDFM)` to `ListMap (GenMap IntMap)`. * Using GenMap offers leaf compression. Which is a trie optimization described by the Note [Compressed TrieMap] in CoreSyn/TrieMap.hs * Using a IntMap removes the overhead associated with UniqDFM. The reasoning why this is deterministic after the change: * IntMap is deterministic given the same keys. * Labels have a Int representation, so for the same Labels we get the same keys, hence the same result for each run. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/15103> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 11