August 2015 - ghc-tickets

Re: [GHC] #5278: System.Random.randomIvalInteger makes invalid assumptions about RandomGen
by GHC 05 Aug '15

05 Aug '15

#5278: System.Random.randomIvalInteger makes invalid assumptions about RandomGen -------------------------------------+------------------------------------- Reporter: rrnewton | Owner: ekmett Type: bug | Status: closed Priority: normal | Milestone: ⊥ Component: Core Libraries | Version: 7.0.3 Resolution: duplicate | Keywords: incorrect | assumption Operating System: Unknown/Multiple | Architecture: Type of failure: Incorrect result | Unknown/Multiple at runtime | Test Case: Blocked By: | Blocking: Related Tickets: #8898 | Differential Revisions: -------------------------------------+------------------------------------- Changes (by thomie): * status: new => closed * resolution: => duplicate * related: => #8898 Comment: This should be fixed now, see #8898. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/5278#comment:3> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] Batch modify: #322, #1307, #3927, #4139, #5728, #5762, #6124, ...
by GHC 05 Aug '15

05 Aug '15

Batch modification to #322, #1307, #3927, #4139, #5728, #5762, #6124, #7669, #8494, #8853, #8970, #9113, #9951 by thomie: Action: resolve Comment: Because the existence of duplicate tickets makes doing a [wiki:BugSweep] of the bug tracker more cumbersome, I'm closing these tickets as duplicate. Don't worry, they're still listed on [wiki:PatternMatchCheck], and will hopefully all be addressed by the [http://git.haskell.org/ghc.git/shortlog/refs/heads/wip/gadtpm work] on #595 ("Overhaul GHC's overlapping/non-exhaustive pattern checking"). * #322 * #1307 * #3927 * #4139 * #5728 * #5762 * #6124 * #7669 * #8494 * #8853 * #8970 * #9113 * #9951 -- Tickets URL: <http://ghc.haskell.org/trac/ghc/query?id=322%2C1307%2C3927%2C4139%2C5728%2C…> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #3575: mkStdGen and split conspire to make some programs predictable
by GHC 05 Aug '15

05 Aug '15

#3575: mkStdGen and split conspire to make some programs predictable -------------------------------------+------------------------------------- Reporter: rwbarton | Owner: ekmett Type: bug | Status: closed Priority: normal | Milestone: ⊥ Component: Core Libraries | Version: 6.10.1 Resolution: duplicate | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Revisions: -------------------------------------+------------------------------------- Changes (by thomie): * status: new => closed * resolution: => duplicate Comment: This is now being tracked in the `random` bug tracker on Github: [https://github.com/haskell/random/issues/25 The seeds generated by split are not independent]. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/3575#comment:10> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #3620: The seeds generated by split are not independent
by GHC 05 Aug '15

05 Aug '15

#3620: The seeds generated by split are not independent -------------------------------------+------------------------------------- Reporter: NickSmallbone | Owner: ekmett Type: bug | Status: closed Priority: normal | Milestone: ⊥ Component: Core Libraries | Version: 6.11 Resolution: duplicate | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Revisions: -------------------------------------+------------------------------------- Changes (by thomie): * status: new => closed * resolution: => duplicate Comment: This is now being tracked in the `random` bug tracker on Github: [https://github.com/haskell/random/issues/25 The seeds generated by split are not independent]. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/3620#comment:6> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #10721: GHC INLINE pragma behavior does not match docs
by GHC 05 Aug '15

05 Aug '15

#10721: GHC INLINE pragma behavior does not match docs -------------------------------------+------------------------------------- Reporter: wyager | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.2 Keywords: | Operating System: MacOS X specialize, inline, inlinable, | pragma | Architecture: x86_64 | Type of failure: Incorrect (amd64) | warning at compile-time Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- Per https://downloads.haskell.org/~ghc/7.8.4/docs/html/users_guide/pragmas.html #specialize-pragma , I should be able to SPECIALIZE an imported function if the function is marked as INLINABLE at the definition site. While trying to debug a related issue (#10720), I tried making the original function `(••)` INLINABLE and then SPECIALIZing it at the call site. However, GHC tells me "`Warning: SPECIALISE pragma on INLINE function probably won't fire: ‘••’`". -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10721> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #10615: Panic; no skolem info (partial type signatures)
by GHC 05 Aug '15

05 Aug '15

#10615: Panic; no skolem info (partial type signatures) -------------------------------------+------------------------------------- Reporter: holzensp | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Blocked By: Test Case: | Related Tickets: Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- This rather minimal program {{{#!hs f :: _ => _f f = const }}} results in this: {{{ GHCi, version 7.10.1: http://www.haskell.org/ghc/ :? for help [1 of 1] Compiling Main ( Panic.hs, interpreted ) Panic.hs:1:6: Found hole ‘_’ with inferred constraints: () To use the inferred type, enable PartialTypeSignatures In the type signature for ‘f’: _ => _f Panic.hs:2:5: Couldn't match expected type ‘_f’ with actual type ‘a0 -> b0 -> a0’ ‘_f’ is untouchable inside the constraints () bound by the inferred type of f :: _f at Panic.hs:2:1-9ghc: panic! (the 'impossible' happened) (GHC version 7.10.1 for x86_64-apple-darwin): No skolem info: _f_alF[sk] }}} The newest open bugs (that reporter could find at the time of reporting) with "no skolem info" in their description are #10404 and #10432. The newest bug (which is closed) is #10503. It seems likely, though, that the above comes from the PartialTypeSignatures implementation (considering the simplicity of the repro). -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10615> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #10724: ghci crashing when using copy/paste in Gnome Terminal on Debian Jessie
by GHC 05 Aug '15

05 Aug '15

#10724: ghci crashing when using copy/paste in Gnome Terminal on Debian Jessie -------------------------------------+------------------------------------- Reporter: martinvlk | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: GHCi | Version: 7.10.2 Keywords: | Operating System: Unknown/Multiple Architecture: x86_64 | Type of failure: None/Unknown (amd64) | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- When I copy/paste text from a ghci instance running in the terminal, quite often it stops responding with the following message: <stdin>: hGetChar: hardware fault (Input/output error) I then have to kill it (Ctrl-C) and restart. It's GHCi, version 7.10.2: http://www.haskell.org/ghc/ :? for help but I remember the same was happening in 7.10.1 too. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10724> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #10704: Can't lookup fixities of infix types in Template Haskell
by GHC 05 Aug '15

05 Aug '15

#10704: Can't lookup fixities of infix types in Template Haskell -------------------------------------+------------------------------------- Reporter: | Owner: RyanGlScott | Type: feature | Status: new request | Priority: normal | Milestone: Component: Template | Version: 7.10.1 Haskell | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- Currently, Template Haskell allows you to reify the fixity of regular functions, typeclass functions, and data constructors: {{{ λ> import Language.Haskell.TH λ> $(reify '($) >>= stringE . show) "VarI GHC.Base.$ (ForallT [KindedTV a_822083586 StarT,KindedTV b_822083587 (ConT GHC.Prim.OpenKind)] [] (AppT (AppT ArrowT (AppT (AppT ArrowT (VarT a_822083586)) (VarT b_822083587))) (AppT (AppT ArrowT (VarT a_822083586)) (VarT b_822083587)))) Nothing (Fixity 0 InfixR)" λ> $(reify '(+) >>= stringE . show) "ClassOpI GHC.Num.+ (ForallT [KindedTV a_1627404054 StarT] [AppT (ConT GHC.Num.Num) (VarT a_1627404054)] (AppT (AppT ArrowT (VarT a_1627404054)) (AppT (AppT ArrowT (VarT a_1627404054)) (VarT a_1627404054)))) GHC.Num.Num (Fixity 6 InfixL)" λ> $(reify '(:%) >>= stringE . show) "DataConI GHC.Real.:% (ForallT [KindedTV a_1627412324 StarT] [] (AppT (AppT ArrowT (VarT a_1627412324)) (AppT (AppT ArrowT (VarT a_1627412324)) (AppT (ConT GHC.Real.Ratio) (VarT a_1627412324))))) GHC.Real.Ratio (Fixity 9 InfixL)" }}} However, you can't do the same for infix typeclasses, type constructors, or type families: {{{ λ> :set -XTypeOperators -XTypeFamilies -XMultiParamTypeClasses λ> class a :=> b; infixr 5 :=> λ> $(reify ''(:=>) >>= stringE . show) "ClassI (ClassD [] Ghci5.:=> [KindedTV a_1627424666 StarT,KindedTV b_1627424667 StarT] [] []) []" λ> type a :+: b = Either a b; infixr 5 :+: λ> $(reify ''(:+:) >>= stringE . show) "TyConI (TySynD Ghci7.:+: [KindedTV a_1627426783 StarT,KindedTV b_1627426784 StarT] (AppT (AppT (ConT Data.Either.Either) (VarT a_1627426783)) (VarT b_1627426784)))" λ> $(reify ''(:*:) >>= stringE . show) "FamilyI (FamilyD TypeFam Ghci9.:*: [KindedTV a_1627426938 StarT,KindedTV b_1627426939 StarT] (Just StarT)) []" }}} We should add a {{{Fixity}}} field to {{{ClassI}}}, {{{TyConI}}}, and {{{FamilyI}}} to make them consistent with the other {{{Info}}} constructors that allow for infix things. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10704> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 13

[GHC] #10706: Make -fcpr-off a dynamic flag
by GHC 05 Aug '15

05 Aug '15

#10706: Make -fcpr-off a dynamic flag -------------------------------------+------------------------------------- Reporter: darchon | Owner: Type: feature | Status: new request | Priority: normal | Milestone: 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 Revisions: | -------------------------------------+------------------------------------- I've run into a situation where I wanted to turn off CPR analysis for a single module in my project, but could not do that because {{{-fcpr-off}}} is a static flag. I will supply a patch that makes {{{-fcpr-off}}} a dynamic flag. From what I understand, this would also bring us one step closer to the desired goal of getting rid of static flags entirely. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10706> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

Re: [GHC] #2747: Excessive Memory Usage: space leak with foldl' on Integer
by GHC 05 Aug '15

05 Aug '15

#2747: Excessive Memory Usage: space leak with foldl' on Integer -------------------------------------+--------------------------------- Reporter: guest | Owner: igloo Type: merge | Status: closed Priority: normal | Milestone: 6.10.2 Component: Runtime System | Version: 6.10.1 Resolution: fixed | Keywords: Operating System: Unknown/Multiple | Architecture: Unknown/Multiple Test Case: | -------------------------------------+--------------------------------- Comment (by Ben Gamari <ben@…>): In [changeset:"dd7e1880ae078a2d9f254dc5d8f330121e0ec291/ghc" dd7e1880/ghc]: {{{ #!CommitTicketReference repository="ghc" revision="dd7e1880ae078a2d9f254dc5d8f330121e0ec291" Add framework flags when linking a dynamic library This fixes the GHC side of trac #10568. So `cabal install --ghc-options="-framework GLUT" GLUT` creates a correctly linked GLUT.dylib. We still need to explictly pass `--ghc-options="-framework GLUT"` because the Cabal side #10568 is not fixed. Update: the Cabal side of #10568 is fixed by [Cabal#2747](https://github.com/haskell/cabal/pull/2747) Test Plan: validate Reviewers: austin, rwbarton, bgamari Reviewed By: bgamari Subscribers: rwbarton, thomie Differential Revision: https://phabricator.haskell.org/D1115 GHC Trac Issues: #10568 }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/2747#comment:9> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0