February 2015 - ghc-tickets

[GHC] #8556: Invalid constructor names are accepted in data declarations
by GHC 23 Mar '15

23 Mar '15

#8556: Invalid constructor names are accepted in data declarations ------------------------------------+------------------------------------- Reporter: dolio | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.6.3 Keywords: | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- Earlier today, someone was asking on #haskell why the constructor name {{{(^)}}} wouldn't work in GADT definitions. My response was that {{{(^)}}} isn't a constructor name, but much to my surprise, GHC accepts such names in a regular data declaration: {{{ data Foo = F | (^^^^) Int Int }}} This creates a {{{Foo}}} type and value constructor {{{F}}}, but no value constructor {{{(^^^^)}}}. However, in 7.6.3, if DataKinds are enabled, both constructors appear at the type level. In HEAD, the same definition is accepted, with only {{{F}}} existing at the value level, as before. But at the type level, both {{{F}}} and {{{(^^^^)}}} just generate errors that {{{Foo}}} is not a promotable type. At that point, I think there's no question that the declaration should just be ruled out. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8556> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #9194: Remove magic parsing of (~)
by GHC 23 Mar '15

23 Mar '15

#9194: Remove magic parsing of (~) ------------------------------------+------------------------------------- Reporter: goldfire | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | 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 seems that `(~)` has magical parsing: {{{ Prelude> :i (~) <interactive>:1:2: parse error on input ‘~’ Prelude> :k (~) (~) :: k -> k -> Constraint Prelude> :i (+) -- for comparison class Num a where (+) :: a -> a -> a ... -- Defined in ‘GHC.Num’ infixl 6 + }}} Now that type operators do not need to start with a colon, this magic seems unnecessary. The `(~)` syntax does not need to be wired-in, unless I'm missing something. (The ''definition'' of `(~)` needs to be wired in -- it's just the syntax that doesn't need to be.) -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9194> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #9849: GHC readme.md seems out of date
by GHC 23 Mar '15

23 Mar '15

#9849: GHC readme.md seems out of date -------------------------------------+------------------------------------- Reporter: carter | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | 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: -------------------------------------+------------------------------------- https://github.com/ghc/ghc/blob/89f5f314e32c3e80c71f4b3dcc8835ae74d7d57f/RE… still mentions using sync-all script, but the wiki docs say its deprecated https://ghc.haskell.org/trac/ghc/wiki/Building/SyncAll -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9849> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

[GHC] #8727: HLearn test on ubuntu Precise x64 within vagrant Box
by GHC 23 Mar '15

23 Mar '15

#8727: HLearn test on ubuntu Precise x64 within vagrant Box ----------------------------------+-------------------------- Reporter: teuffy | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.6.3 Keywords: | Operating System: Linux Architecture: x86_64 (amd64) | Type of failure: Other Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ----------------------------------+-------------------------- ghc: panic! (the 'impossible' happened) (GHC version 7.6.3 for x86_64-unknown-linux): loadArchive "/home/vagrant/.cabal/lib/x86_64-linux- ghc-7.6.3/HLearn-2.0.0.0/libHSHLearn-2.0.0.0.a": failed {{{ import HLearn.Algebra import Control.DeepSeq import HLearn.Models.Distributions import Statistics.Distribution.Normal dataset = [1,2,3,4,5,6] --dist = train dataset :: Normal Double data Marble = Red | Green | Blue | White deriving (Read,Show,Eq,Ord) simonBag :: [Marble] simonBag = [Red, Red, Red, Green, Blue, Green, Red, Blue, Green, Green, Red, Red, Blue, Red, Red, Red, White] simonDist = train simonBag :: Categorical Double Marble }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8727> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 18

[GHC] #8078: Core lint failure for profiled code
by GHC 23 Mar '15

23 Mar '15

#8078: Core lint failure for profiled code -----------------------------------+--------------------------------------- Reporter: ezyang | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.7 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Compile-time crash Unknown/Multiple | Test Case: T4492 Difficulty: Unknown | Blocking: Blocked By: | Related Tickets: | -----------------------------------+--------------------------------------- Probably related to the cardinality changes. Here is the dump: =====> T4492(profasm) 3361 of 3686 [2, 51, 0] cd ./indexed-types/should_compile && '/srv/code/ghc-build- sandbox/inplace/bin/ghc-stage2' -fforce-recomp -dcore-lint -dcmm-lint -dno-debug-output -no-user-package-db -rtsopts -fno-ghci-history -c T4492.hs -O -prof -static -auto-all >T4492.comp.stderr 2>&1 Compile failed (status 256) errors were: *** Core Lint errors : in result of CorePrep *** {-# LINE 12 "T4492.hs #-}: Warning: [RHS of T4492.$fCMaybe_$cgo :: forall a_aex. T4492.C a_aex => forall b_ae0. (forall a1_ae1. Data.Maybe.Maybe a1_ae1 -> b_ae0 -> a1_ae1) -> Data.Maybe.Maybe a_aex -> T4492.F (Data.Maybe.Maybe a_aex) b_ae0] Demand type has 4 arguments, rhs has 3 arguments, T4492.$fCMaybe_$cgo Binder's strictness signature: DmdType <C(C(S)),1*C1(C1(U))><L,U><L,U><L,U> *** Offending Program *** T4492.go [InlPrag=INLINE] :: forall a_adZ. T4492.C a_adZ => forall b_ae0. (forall a_ae1. Data.Maybe.Maybe a_ae1 -> b_ae0 -> a_ae1) -> a_adZ -> T4492.F a_adZ b_ae0 [GblId[ClassOp], Arity=1, Caf=NoCafRefs, Str=DmdType <S,U>, Unf=OtherCon []] T4492.go = \ (@ a_adZ) (tpl_sgH [Occ=Once] :: T4492.C a_adZ) -> tpl_sgH `cast` (T4492.NTCo:C[0] <a_adZ> :: T4492.C a_adZ ~# (forall b_ae0. (forall a_ae1. Data.Maybe.Maybe a_ae1 -> b_ae0 -> a_ae1) -> a_adZ -> T4492.F a_adZ b_ae0)) T4492.$fCMaybe1 :: forall a_aex. T4492.C a_aex => forall b_afH. (forall a1_ae1. Data.Maybe.Maybe a1_ae1 -> b_afH -> a1_ae1) -> Data.Maybe.Maybe a_aex -> b_afH -> T4492.F a_aex b_afH [GblId, Arity=4, Caf=NoCafRefs, Str=DmdType <C(C(S)),1*C1(C1(U))><L,U><L,U><L,U>, Unf=OtherCon []] T4492.$fCMaybe1 = \ (@ a_aex) ($dC_sgI [Occ=OnceL] :: T4492.C a_aex) (@ b_afH) (f_sgJ :: forall a1_ae1. Data.Maybe.Maybe a1_ae1 -> b_afH -> a1_ae1) (a1_sgK [Occ=OnceL] :: Data.Maybe.Maybe a_aex) (b1_sgL [Occ=OnceL] :: b_afH) -> scctick<go> let { sat_sgM [Occ=Once] :: a_aex [LclId, Str=DmdType] sat_sgM = f_sgJ @ a_aex a1_sgK b1_sgL } in ($dC_sgI `cast` (T4492.NTCo:C[0] <a_aex> :: T4492.C a_aex ~# (forall b_ae0. (forall a_ae1. Data.Maybe.Maybe a_ae1 -> b_ae0 -> a_ae1) -> a_aex -> T4492.F a_aex b_ae0))) @ b_afH f_sgJ sat_sgM T4492.$fCMaybe_$cgo :: forall a_aex. T4492.C a_aex => forall b_ae0. (forall a1_ae1. Data.Maybe.Maybe a1_ae1 -> b_ae0 -> a1_ae1) -> Data.Maybe.Maybe a_aex -> T4492.F (Data.Maybe.Maybe a_aex) b_ae0 [GblId, Arity=3, Caf=NoCafRefs, Str=DmdType <C(C(S)),1*C1(C1(U))><L,U><L,U><L,U>, Unf=OtherCon []] T4492.$fCMaybe_$cgo = (\ (@ a_aex) (eta_B3 [Occ=Once] :: T4492.C a_aex) (@ b_afH) (eta_B2 [Occ=Once] :: forall a1_ae1. Data.Maybe.Maybe a1_ae1 -> b_afH -> a1_ae1) (eta_B1 [Occ=Once] :: Data.Maybe.Maybe a_aex) -> T4492.$fCMaybe1 @ a_aex eta_B3 @ b_afH eta_B2 eta_B1) `cast` (forall a_aex. <T4492.C a_aex> -> forall b_afH. <forall a1_ae1. Data.Maybe.Maybe a1_ae1 -> b_afH -> a1_ae1> -> <Data.Maybe.Maybe a_aex> -> Sym (T4492.TFCo:R:FMaybeb[0] <a_aex> <b_afH>) :: (forall a_aex. T4492.C a_aex => forall b_afH. (forall a1_ae1. Data.Maybe.Maybe a1_ae1 -> b_afH -> a1_ae1) -> Data.Maybe.Maybe a_aex -> b_afH -> T4492.F a_aex b_afH) ~# (forall a_aex. T4492.C a_aex => forall b_afH. (forall a1_ae1. Data.Maybe.Maybe a1_ae1 -> b_afH -> a1_ae1) -> Data.Maybe.Maybe a_aex -> T4492.F (Data.Maybe.Maybe a_aex) b_afH)) T4492.$fCMaybe [InlPrag=INLINE (sat-args=0)] :: forall a_aex. T4492.C a_aex => T4492.C (Data.Maybe.Maybe a_aex) [GblId[DFunId(nt)], Arity=3, Caf=NoCafRefs, Str=DmdType <C(C(S)),1*C1(C1(U))><L,U><L,U><L,U>, Unf=OtherCon []] T4492.$fCMaybe = (\ (@ a_aex) (eta_B3 [Occ=Once] :: T4492.C a_aex) (@ b_afH) (eta_B2 [Occ=Once] :: forall a1_ae1. Data.Maybe.Maybe a1_ae1 -> b_afH -> a1_ae1) (eta_B1 [Occ=Once] :: Data.Maybe.Maybe a_aex) -> T4492.$fCMaybe1 @ a_aex eta_B3 @ b_afH eta_B2 eta_B1) `cast` (forall a_aex. <T4492.C a_aex> -> (forall b_afH. <forall a1_ae1. Data.Maybe.Maybe a1_ae1 -> b_afH -> a1_ae1> -> <Data.Maybe.Maybe a_aex> -> Sym (T4492.TFCo:R:FMaybeb[0] <a_aex> <b_afH>)) ; Sym (T4492.NTCo:C[0] <Data.Maybe.Maybe a_aex>) :: (forall a_aex. T4492.C a_aex => forall b_afH. (forall a1_ae1. Data.Maybe.Maybe a1_ae1 -> b_afH -> a1_ae1) -> Data.Maybe.Maybe a_aex -> b_afH -> T4492.F a_aex b_afH) ~# (forall a_aex. T4492.C a_aex => T4492.C (Data.Maybe.Maybe a_aex))) *** End of Offense *** <no location info>: Compilation had errors *** unexpected failure for T4492(profasm) -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8078> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #9261: -S prints incorrect number of bound tasks
by GHC 22 Mar '15

22 Mar '15

#9261: -S prints incorrect number of bound tasks ------------------------------------+------------------------------------- Reporter: edsko | Owner: Type: bug | Status: new Priority: low | Milestone: Component: Compiler | 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: | ------------------------------------+------------------------------------- In `Stats.c` we have: {{{ statsPrintf(" TASKS: %d (%d bound, %d peak workers (%d total), using -N%d)\n", taskCount, taskCount - workerCount, peakWorkerCount, workerCount, n_capabilities); }}} but I think `taskCount - workerCount` must be wrong, because `taskCount` is the _current_ number of tasks, while `workerAcount` is the _total_ number of workers (accumulating). I think it should be: {{{ statsPrintf(" TASKS: %d (%d bound, %d peak workers (%d total), using -N%d)\n", taskCount, taskCount - currentWorkerCount, peakWorkerCount, workerCount, n_capabilities); }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9261> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #8379: sync-all broken when using the GitHub mirror
by GHC 21 Mar '15

21 Mar '15

#8379: sync-all broken when using the GitHub mirror ------------------------------------+------------------------------------- Reporter: tibbe | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Build System | Version: 7.7 Keywords: | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- Making a fresh checkout using the GitHub mirror: {{{ $ git clone git@github.com:tibbe/ghc.git $ ./sync-all get == running git config core.ignorecase true == running git clone git@github.com:tibbe/ghc-tarballs.git ghc-tarballs Cloning into 'ghc-tarballs'... ERROR: Repository not found. fatal: The remote end hung up unexpectedly git failed: 32768 at ./sync-all line 178. == Checking for old haddock repo == Checking for old binary repo == Checking for old mtl repo == Checking for old Cabal repo == Checking for old time from tarball ============================ ATTENTION! You have an old time package in your GHC tree! Please remove it (e.g. "rm -r libraries/time"), and then run "./sync-all get" to get the new repository. ============================ }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8379> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 13

[GHC] #8150: Foreign export requires redundant type signature
by GHC 19 Mar '15

19 Mar '15

#8150: Foreign export requires redundant type signature -------------------------+------------------------------------------------- Reporter: | Owner: evincarofautumn | Status: new Type: bug | Milestone: Priority: | Version: 7.6.3 normal | Operating System: Unknown/Multiple Component: | Type of failure: Incorrect warning at Compiler | compile-time Keywords: | Test Case: Architecture: | Blocking: Unknown/Multiple | Difficulty: | Unknown | Blocked By: | Related Tickets: | -------------------------+------------------------------------------------- With a source file such as: {{{ module FFITest where foreign export ccall "function" function :: IO () function :: IO () function = putStrLn "Hello, world!" }}} The second `function :: IO ()` signature is required to satisfy `-fwarn- missing-signatures`, even though a signature is already present in the `foreign export` declaration, and the types are required to be the same. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8150> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #10011: The Data instance for Ratio violates internal invariants.
by GHC 19 Mar '15

19 Mar '15

#10011: The Data instance for Ratio violates internal invariants. -------------------------------------+------------------------------------- Reporter: ekmett | Owner: ekmett Type: bug | Status: new Priority: normal | Milestone: Component: Core | Version: 7.10.1-rc1 Libraries | Operating System: Unknown/Multiple Keywords: | Type of failure: None/Unknown Architecture: | Blocked By: Unknown/Multiple | Related Tickets: Test Case: | Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- I found this when Simon was cleaning up unused dependencies in {{{ https://phabricator.haskell.org/rGHCc409b6f30373535b6eed92e55d4695688d32be9… }}} The Data instance for Ratio just uses the raw `(:%)` constructor and doesn't check that the result is reduced to normal form. It strikes me that the fix is to add back the Integral constraint on the Data instance and to use `(%)` rather than `(:%)` in the `gfoldl` and `gunfold` code. This restores the invariant and matches the behavior of "virtual constructors" we've used to patch up such problems elsewhere. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10011> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 17

[GHC] #9572: nofib target for just building should be part of validate
by GHC 19 Mar '15

19 Mar '15

#9572: nofib target for just building should be part of validate -------------------------------------+------------------------------------- Reporter: ezyang | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: NoFib benchmark | Version: 7.8.2 suite | Operating System: Keywords: | Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: Difficulty: Unknown | None/Unknown Blocked By: | Test Case: Related Tickets: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- nofib is never run during validates, which makes it easy to bitrot (c.f. AMP changes). It would be better if: 1. We had a build target for just building nofib tests, not running them, (maybe optionally with optimizations turned off, or just `-fno-code`) and 2. This build target was executed during validates This should strike a balance between wanting to avoid having to spend a lot of time running the benchmarks on validate, while still exercising the nofib tests to some degree. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9572> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2