November 2017 - ghc-tickets

Re: [GHC] #459: Bad parse error message
by GHC 15 Nov '17

15 Nov '17

#459: Bad parse error message --------------------------------------+--------------------------------- Reporter: nobody | Owner: (none) Type: bug | Status: closed Priority: normal | Milestone: ⊥ Component: Compiler (Parser) | Version: 6.4.1 Resolution: fixed | Keywords: Operating System: Unknown/Multiple | Architecture: Unknown/Multiple Type of failure: Other | Test Case: Blocked By: | Blocking: Related Tickets: | --------------------------------------+--------------------------------- Comment (by Andrey Mokhov <andrey.mokhov@…>): In [changeset:"c1fcd9b3f60e8420dd228cd4e3efeb9cfa793aa5/ghc" c1fcd9b/ghc]: {{{ #!CommitTicketReference repository="ghc" revision="c1fcd9b3f60e8420dd228cd4e3efeb9cfa793aa5" Squashed 'hadrian/' changes from 5ebb69a..fa3771f fa3771f hadrian: Disable -Wno-undef in files which include ffi.h (#459) f15e851 Do not run configure by default (#458) 5baa8db Fix AppVeyor cache failure (#456) 94dbe9d Fix ghc-cabal build (#455) a679764 Fix CI scripts (#454) 06ec241 Widen bounds on Cabal (#452) git-subtree-dir: hadrian git-subtree-split: fa3771fe6baf5008a8506fec48220f8347ac59af }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/459#comment:13> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #14445: GC timing stats (e.g. mutator_elapsed_ns) decrease over time
by GHC 15 Nov '17

15 Nov '17

#14445: GC timing stats (e.g. mutator_elapsed_ns) decrease over time -------------------------------------+------------------------------------- Reporter: harendra | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Runtime | Version: 8.2.1 System | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Debugging Unknown/Multiple | information is incorrect Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- See this measurement code here in the `gauge` benchmarking tool, which is a clone of the criterion benchmarking tool. : https://github.com/harendra-kumar/hs- gauge/blob/f8c9bcd6ee03c3090f110664ff279a20f2da9130/Gauge/Measurement.hs#L315 . This code displays debug prints when the end time is lower than the start time for `mutator_elapsed_ns`. When we run this benchmark defined in https://github.com/harendra-kumar /hs-gauge/blob/f8c9bcd6ee03c3090f110664ff279a20f2da9130/benchs/GCBug.hs the following output is produced: {{{ $ stack bench --benchmark-arguments "--quick" gauge-0.1.3: benchmarks Running 1 benchmarks... Benchmark gcbug: RUNNING... benchmarking streamingDecreased by: 25818791 : s = 59840178 e = 34021387 Decreased by: 22707134 : s = 76706673 e = 53999539 Decreased by: 4776100 : s = 96201146 e = 91425046 streaming time 50.00 ms benchmarking pipesDecreased by: 12487955 : s = 153949495 e = 141461540 pipes time 50.00 ms benchmarking conduitDecreased by: 5523916 : s = 310913002 e = 305389086 conduit time 50.00 ms Benchmark gcbug: FINISH }}} Notice the "Decreased by" getting printed quite a few times. I have found it to be more likely for these particular libraries i.e. streaming, pipes and conduit; so I have used only these in the benchmarking file `GCBug.hs`. I tried some other streaming libraries as well but it did not occur often in those cases. In these cases it occurs almost every time. I am running it on a macbook. The repository is here: https://github.com/harendra-kumar/hs-gauge/tree /gc-bug . The branch in the repo is `gc-bug`. You can run the benchmark named `gcbug` using the command `stack bench --benchmark-arguments "-- quick"` to reproduce the output above. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14445> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #14467: Support HasCallStack for calls to panic
by GHC 15 Nov '17

15 Nov '17

#14467: Support HasCallStack for calls to panic -------------------------------------+------------------------------------- Reporter: AndreasK | Owner: (none) Type: feature | Status: new request | Priority: low | Milestone: Component: GHC API | Version: 8.2.1 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: -------------------------------------+------------------------------------- As far as I can tell `panic` only reports cost center callstacks currently. It would be useful to include information provided by HasCallStack if available as well. When using the GHC API (or hacking on GHC itself) this would be a convenient way to get more information about what lead to an error that triggers `panic` in a different part of the compiler. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14467> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

[GHC] #14469: Rebuilding profiled stage2 after building stage3 is broken
by GHC 15 Nov '17

15 Nov '17

#14469: Rebuilding profiled stage2 after building stage3 is broken -------------------------------------+------------------------------------- Reporter: bgamari | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: 8.4.1 Component: Build System | Version: 8.2.1 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: -------------------------------------+------------------------------------- The following appears to break, {{{ $ cat <<EOF >mk/build.mk BuildFlavour = prof ifneq "$(BuildFlavour)" "" include mk/flavours/$(BuildFlavour).mk endif EOF $ make stage=2 $ make stage=3 $ touch compiler/stage2/build/Simplify.* $ make stage=2 ... ghc/GHCi/UI/Info.hs:43:1: error: Could not find module ‘TcHsSyn’ Perhaps you haven't installed the profiling libraries for package ‘ghc-8.3’? Locations searched: ghc/./TcHsSyn.p_hi ghc/./TcHsSyn.p_hi-boot ghc/stage2/build/TcHsSyn.p_hi ghc/stage2/build/TcHsSyn.p_hi-boot ghc/stage2/build/ghc/autogen/TcHsSyn.p_hi ghc/stage2/build/ghc/autogen/TcHsSyn.p_hi-boot /mnt/work/ghc/ghc/compiler/stage3/build/TcHsSyn.p_hi | 43 | import TcHsSyn | ^^^^^^^^^^^^^^^^^^^^^^^^ }}} The problem appears to be that the build system registers the stage3 `ghc`, which we didn't build in the profiled way, overwriting the stage2 registration. Hopefully Hadrian will do better here. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14469> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #14450: GHCi spins forever
by GHC 15 Nov '17

15 Nov '17

#14450: GHCi spins forever -------------------------------------+------------------------------------- Reporter: Iceland_jack | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.2.1 Keywords: TypeInType, | Operating System: Unknown/Multiple PolyKinds | Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- The following code compiles just fine (8.3.20170920) {{{#!hs {-# Language KindSignatures, TypeOperators, PolyKinds, TypeOperators, ConstraintKinds, TypeFamilies, DataKinds, TypeInType, GADTs, AllowAmbiguousTypes, InstanceSigs #-} import Data.Kind data TyFun :: Type -> Type -> Type type a ~> b = TyFun a b -> Type type Cat ob = ob -> ob -> Type type SameKind (a :: k) (b :: k) = (() :: Constraint) type family Apply (f :: a ~> b) (x :: a) :: b where Apply IddSym0 x = Idd x class Varpi (f :: i ~> j) where type Dom (f :: i ~> j) :: Cat i type Cod (f :: i ~> j) :: Cat j varpa :: Dom f a a' -> Cod f (Apply f a) (Apply f a') type family Idd (a::k) :: k where Idd (a::k) = a data IddSym0 :: k ~> k where IddSym0KindInference :: IddSym0 l instance Varpi (IddSym0 :: Type ~> Type) where type Dom (IddSym0 :: Type ~> Type) = (->) type Cod (IddSym0 :: Type ~> Type) = (->) varpa :: (a -> a') -> (a -> a') varpa = id }}} But if you change the final instance to {{{#!hs instance Varpi (IddSym0 :: k ~> k) where type Dom (IddSym0 :: Type ~> Type) = (->) }}} it sends GHC for a spin. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14450> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 10

[GHC] #14464: Adjust AltCon Ord instance to match linter requirements.
by GHC 15 Nov '17

15 Nov '17

#14464: Adjust AltCon Ord instance to match linter requirements. -------------------------------------+------------------------------------- Reporter: AndreasK | Owner: (none) Type: task | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.2.1 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: -------------------------------------+------------------------------------- There is a Ord instance for AltCon which is incompatible with the Core Linters requirements. Updating it to match the linters requirement seems useful as it allows easy sorting of alternatives. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14464> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #14353: PowerPC: HEAD validate fails due to warnings in libffi
by GHC 15 Nov '17

15 Nov '17

#14353: PowerPC: HEAD validate fails due to warnings in libffi --------------------------------+---------------------------------------- Reporter: trommler | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.3 Keywords: | Operating System: Unknown/Multiple Architecture: powerpc | Type of failure: Building GHC failed Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: --------------------------------+---------------------------------------- Validate produces the following error message: {{{ In file included from rts/dist/build/ffi.h:58:0: error: 0, from rts/Adjustor.c:46: rts/dist/build/ffitarget.h:157:7: error: error: "FFI_TYPE_LAST" is not defined [-Werror=undef] #if !(FFI_TYPE_LAST == FFI_PPC_TYPE_LAST \ }}} and a few more like this. Perhaps, we could turn off the warning for files that include `ffi.h` on PowerPC. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14353> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #14455: PPC64: Wrong output in print022
by GHC 15 Nov '17

15 Nov '17

#14455: PPC64: Wrong output in print022 -------------------------------------+------------------------------------- Reporter: trommler | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: GHCi | Version: 8.3 Keywords: big endian | Operating System: Unknown/Multiple Architecture: powerpc64 | Type of failure: Incorrect result Test Case: | at runtime ghci.debugger/scripts/print022 | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- On powerpc64 Linux print022 fails with the following: {{{ -test = C 1 32 1.2 1.23 'x' 1 1.2 1.23 +test = C 1 32 0.0 1.23 'x' 1 1.2 1.23 }}} Note, the fourth item is 0.0 instead of 1.2. In ticket #11262 the fourth item was still correct. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14455> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #14466: Error while installing codeworld-api
by GHC 15 Nov '17

15 Nov '17

#14466: Error while installing codeworld-api -------------------------------------+------------------------------------- Reporter: Zalathorm | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.2.1 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: -------------------------------------+------------------------------------- C:\Users\****>cabal install codeworld-api Resolving dependencies... Configuring codeworld-api-0.2.1.0... Building codeworld-api-0.2.1.0... Failed to install codeworld-api-0.2.1.0 Build log ( C:\Users\****\AppData\Roaming\cabal\logs\ghc-8.2.1\codeworld- api-0.2.1.0-FP1wjOCq6YC5GIs4ve2sRp.log ): Preprocessing library for codeworld-api-0.2.1.0.. Building library for codeworld-api-0.2.1.0.. [1 of 6] Compiling CodeWorld.Color ( src\CodeWorld\Color.hs, dist\build\CodeWorld\Color.o ) [2 of 6] Compiling CodeWorld.Picture ( src\CodeWorld\Picture.hs, dist\build\CodeWorld\Picture.o ) [3 of 6] Compiling CodeWorld.Event ( src\CodeWorld\Event.hs, dist\build\CodeWorld\Event.o ) [4 of 6] Compiling CodeWorld.CollaborationUI ( src\CodeWorld\CollaborationUI.hs, dist\build\CodeWorld\CollaborationUI.o ) ghc.exe: panic! (the 'impossible' happened) (GHC version 8.2.1 for x86_64-unknown-mingw32): runtimeRepPrimRep typePrimRep (r_abQP :: TYPE rep_abQO) rep_abQO Call stack: CallStack (from HasCallStack): prettyCurrentCallStack, called at compiler\utils\Outputable.hs:1133:58 in ghc:Outputable callStackDoc, called at compiler\utils\Outputable.hs:1137:37 in ghc:Outputable pprPanic, called at compiler\simplStg\RepType.hs:360:5 in ghc:RepType Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug cabal: Leaving directory 'C:\Users\****\AppData\Local\Temp\cabal-tmp-5456 \codeworld-api-0.2.1.0' cabal: Error: some packages failed to install: codeworld-api-0.2.1.0-FP1wjOCq6YC5GIs4ve2sRp failed during the building phase. The exception was: ExitFailure 1 -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14466> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #14463: Pattern synonym for appliation
by GHC 14 Nov '17

14 Nov '17

#14463: Pattern synonym for appliation -------------------------------------+------------------------------------- Reporter: Iceland_jack | Owner: (none) Type: feature | Status: new request | Priority: normal | Milestone: Component: Compiler | Version: 8.2.1 Keywords: | Operating System: Unknown/Multiple PatternSynonyms | Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Does this make sense {{{#!hs pattern f :$ x = f x }}} where `Just :$ False` is the same as the pattern `Just False`. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14463> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2