July 2016 - ghc-tickets

[GHC] #12026: Pattern match failure in RnNames.hs
by GHC 12 Jul '16

12 Jul '16

#12026: Pattern match failure in RnNames.hs -------------------------------------+------------------------------------- Reporter: davean | Owner: Type: bug | Status: new Priority: high | Milestone: Component: Compiler | Version: 8.0.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Compile-time Unknown/Multiple | crash Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- When compiling dimensional-1.0.1.1 after patching numtype-dk with the patch at https://github.com/bjornbm/numtype-dk/issues/12 GHC does the "impossible": [ 3 of 16] Compiling Numeric.Units.Dimensional.Dimensions.TypeLevel ( src/Numeric/Units/Dimensional/Dimensions/TypeLevel.hs, dist/dist-sandbox- 68389a1a/build/Numeric/Units/Dimensional/Dimensions/TypeLevel.o ) ghc: panic! (the 'impossible' happened) (GHC version 8.0.0.20160421 for x86_64-unknown-linux): Pattern match failure in do expression at compiler/rename/RnNames.hs:902:12-50 -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12026> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 8

[GHC] #12094: Unlifted types and pattern synonym signatures
by GHC 12 Jul '16

12 Jul '16

#12094: Unlifted types and pattern synonym signatures -------------------------------------+------------------------------------- Reporter: akio | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 (Type checker) | Keywords: pattern | Operating System: Unknown/Multiple synonyms | Architecture: | Type of failure: GHC rejects Unknown/Multiple | valid program Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- The following program is rejected by ghc-8.0.1: {{{#!hs {-# LANGUAGE MagicHash #-} {-# LANGUAGE PatternSynonyms #-} module Foo where import GHC.Exts (Int#) pattern Zero :: Int# -- commenting out this line works pattern Zero <- 0# }}} {{{ % ghc-stage2 unlifted-pattern-synonym.hs [1 of 1] Compiling Foo ( unlifted-pattern-synonym.hs, unlifted-pattern-synonym.o ) unlifted-pattern-synonym.hs:7:17: error: • Expecting a lifted type, but ‘Int#’ is unlifted • In the type ‘Int#’ }}} Commenting out the type signature removes the error. GHC 7.10.3 accepts this program, with and without the type signature. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12094> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #11690: stage 1 compiler silently ignores plugins
by GHC 12 Jul '16

12 Jul '16

#11690: stage 1 compiler silently ignores plugins -------------------------------------+------------------------------------- Reporter: osa1 | Owner: osa1 Type: feature | Status: new request | Priority: lowest | Milestone: Component: Compiler | Version: 7.10.3 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: -------------------------------------+------------------------------------- It can't load plugins but it shouldn't just silently ignore IMHO. At least a warning would be nice. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11690> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #12212: GHC 8.0.1 crash
by GHC 12 Jul '16

12 Jul '16

#12212: GHC 8.0.1 crash -------------------------------------+------------------------------------- Reporter: dibblego | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 Keywords: | Operating System: Linux Architecture: | Type of failure: Compile-time Unknown/Multiple | crash Test Case: | Blocked By: Blocking: | Related Tickets: 9160, 10602 Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Attempting to install the geodetics package results in a GHC 8.0.1 crash. {{{ % ghc --version The Glorious Glasgow Haskell Compilation System, version 8.0.1 % uname -a Linux moron 4.6.2-1-ARCH #1 SMP PREEMPT Wed Jun 8 08:40:59 CEST 2016 x86_64 GNU/Linux % cabal install geodetics-0.0.4 ...omitted [2 of 9] Compiling Geodetics.Ellipsoids ( src/Geodetics/Ellipsoids.hs, dist/dist-sandbox-49e45c7e/build/Geodetics/Ellipsoids.o ) ghc: panic! (the 'impossible' happened) (GHC version 8.0.1 for x86_64-unknown-linux): Template variable unbound in rewrite rule Variable: cobox_scSX Rule "SPEC t @ a @ 'DQuantity @ 'DQuantity @ d @ d' @ 'DQuantity @ 'DQuantity @ d @ d' @ 'DQuantity @ 'DQuantity @ d @ d'" Rule bndrs: [cobox_scSX, cobox_scSY, cobox_scSZ, cobox_scT0, cobox_scT1, $dKnownVariant_scT2, $dKnownVariant_scT3, $dKnownVariant_scT4, $dKnownVariant_scT5, $dKnownVariant_scT6, $dKnownVariant_scT7, $dNum_scT8] LHS args: [TYPE: a_ac0a, TYPE: 'DQuantity, TYPE: 'DQuantity, TYPE: d_ac0b, TYPE: d'_ac0c, TYPE: 'DQuantity, TYPE: 'DQuantity, TYPE: d_ac0b, TYPE: d'_ac0c, TYPE: 'DQuantity, TYPE: 'DQuantity, TYPE: d_ac0b, TYPE: d'_ac0c, CO: <d_ac0b * d'_ac0c>_N, CO: cobox_scSY, CO: <d_ac0b * d'_ac0c>_N, CO: cobox_scT0, CO: cobox_scT1, $dKnownVariant_scT2, $dKnownVariant_scT3, $dKnownVariant_scT4, $dKnownVariant_scT5, $dKnownVariant_scT6, $dKnownVariant_scT7, $dNum_scT8] Actual args: [TYPE: a_ac0a, TYPE: 'DQuantity, TYPE: 'DQuantity, TYPE: d_ac0b, TYPE: d'_ac0c, TYPE: 'DQuantity, TYPE: 'DQuantity, TYPE: d_ac0b, TYPE: d'_ac0c, TYPE: 'DQuantity, TYPE: 'DQuantity, TYPE: d_ac0b, TYPE: d'_ac0c, CO: <d_ac0b * d'_ac0c>_N, CO: D:R:*[1], CO: <d_ac0b * d'_ac0c>_N, CO: D:R:*[1], CO: D:R:*[1], $fKnownVariantDQuantity, $fKnownVariantDQuantity, $fKnownVariantDQuantity, $fKnownVariantDQuantity, $fKnownVariantDQuantity, $fKnownVariantDQuantity, $dNum_acqm, tx_aaQW, v_aaQZ] Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug cabal: Error: some packages failed to install: geodetics-0.0.4 failed during the building phase. The exception was: ExitFailure 1 }}} Similarity to #10602 and #9160 has been pointed out. I also tried: {{{ % cabal --ghc-options="-O2" install geodetics-0.0.4 }}} with no difference in the crash outcome. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12212> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

Re: [GHC] #393: functions without implementations
by GHC 12 Jul '16

12 Jul '16

#393: functions without implementations -------------------------------------+------------------------------------- Reporter: c_maeder | Owner: Type: feature request | Status: new Priority: normal | Milestone: ⊥ Component: Compiler (Type | Version: None checker) | Resolution: | Keywords: newcomer 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 nomeata): * status: closed => new * resolution: wontfix => Comment: I find Iceland_jack’s examples convincing, and this is a fun nice newcomer task, hence reopening. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/393#comment:32> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #12168: panic! (the 'impossible' happened) with gi-gtk 3.0.4
by GHC 12 Jul '16

12 Jul '16

#12168: panic! (the 'impossible' happened) with gi-gtk 3.0.4 -------------------------------------+------------------------------------- Reporter: drb226 | 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: -------------------------------------+------------------------------------- I'm able to get two different panics, depending on how I try to build gi- gtk. Firstly, on my mac: {{{ brew install gtk+3 stack unpack gi-gtk-3.0.4 cd gi-gtk-3.0.4 stack init --resolver nightly-2016-06-08 PKG_CONFIG_PATH=/usr/local/opt/libffi/lib/pkgconfig/ stack build }}} This results in {{{ [650 of 650] Compiling GI.Gtk ( GI/Gtk.hs, .stack- work/dist/x86_64-osx/Cabal-1.24.0.0/build/GI/Gtk.o ) ghc: panic! (the 'impossible' happened) (GHC version 8.0.1 for x86_64-apple-darwin): RnEnv.plusAvail bindingSignalWriteNext CheckButtonK{CheckButtonK} }}} Secondly, also on my mac, but within an ubuntu-based docker image: {{{ docker run -it snoyberg/stackage:nightly cd ~ stack unpack gi-gtk-3.0.4 cd gi-gtk-3.0.4 stack init --resolver nightly-2016-06-08 stack build }}} This results in {{{ [599 of 654] Compiling GI.Gtk.Objects.Widget ( GI/Gtk/Objects/Widget.hs, .stack-work/dist/x86_64-linux/Cabal-1.24.0.0/build/GI/Gtk/Objects/Widget.o ) ghc: panic! (the 'impossible' happened) (GHC version 8.0.1 for x86_64-unknown-linux): urk! lookup local fingerprint $fAttrInfoWidgetMarginTopPropertyInfo4 }}} The latter panic was first observed on the stackage build server, during the June 8th nightly build. That's why, to use the same dependencies, I specified the nightly-2016-06-08 stack resolver in order to reproduce it. I originally recorded it here: https://github.com/fpco/stackage/issues/1569 -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12168> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

Re: [GHC] #393: functions without implementations
by GHC 11 Jul '16

11 Jul '16

#393: functions without implementations -------------------------------------+------------------------------------- Reporter: c_maeder | Owner: Type: feature request | Status: closed Priority: normal | Milestone: ⊥ Component: Compiler (Type | Version: None checker) | Resolution: wontfix | Keywords: newcomer Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by Iceland_jack): Useful when inferring types in GHCi. (ticket:2256#comment:25) I want to be able to write {{{ >>> data A; data B a; data D a; data S a; data Z >>> k1 :: (A -> B x) -> D (S x) >>> k2 :: A -> D x -> B (S x) >>> k3 :: D Z }}} {{{ >>> :t k1 (\a1 -> k2 a1 (k1 (\a2 -> k2 a1 k3))) k1 (\a1 -> k2 a1 (k1 (\a2 -> k2 a1 k3))) :: D (S (S (S (S Z)))) >>> :t k1 (\a1 -> k2 a1 (k1 (\a2 -> k2 a2 k3))) k1 (\a1 -> k2 a1 (k1 (\a2 -> k2 a2 k3))) :: D (S (S (S (S Z)))) }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/393#comment:31> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #2256: Incompleteness of type inference: must quantify over implication constraints
by GHC 11 Jul '16

11 Jul '16

#2256: Incompleteness of type inference: must quantify over implication constraints -------------------------------------+------------------------------------- Reporter: simonpj | Owner: simonpj Type: bug | Status: new Priority: lowest | Milestone: Component: Compiler (Type | Version: 6.8.2 checker) | Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: Type of failure: GHC rejects | Unknown/Multiple valid program | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by Iceland_jack): From [http://arxiv.org/pdf/1511.09394v1.pdf Proof Relevant Corecursive Resolution], I will copy the relevant text: > Derivable type classes.::: > > Hinze and Peyton Jones wanted to use an instance of the form > > {{{#!hs > instance (Binary a, Binary (f (GRose f a))) => Binary (GRose f a) > }}} > > but discovered that it causes resolution to diverge. They suggested the following as a replacement: > > {{{#!hs > instance (Binary a, ∀b. Binary b => Binary (f b)) => Binary (GRose f a) > }}} > > Unfortunately, Haskell does not support instances with ''polymorphic higher-order instance contexts''. Nevertheless, allowing such implication constraints would greatly increase the expressivity of corecursive resolution. In the terminology of our paper, it amounts to extending Horn formulas to intuitionistic formulas. Working with intuitionistic formulas would require a certain amount of searching, as the non-overlapping condition for Horn formulas is not enough to ensure uniqueness of the evidence. For example, consider the following axioms: > > {{{ > κ₁ : (A ⇒ B x) ⇒ D (S x) > ĸ₂ : A, D x ⇒ B (S x) > κ₃ : ⇒ D Z > }}} > We have two distinct proof terms for `D (S (S (S (S Z)))))`: > > {{{ > κ₁ (λα₁. ĸ₂ α₁ (ĸ₁ (λα₂. ĸ₂ α₁ ĸ₃)) > κ₁ (λα₁. ĸ₂ α₁ (ĸ₁ (λα₂. ĸ₂ α₂ ĸ₃)) > }}} > > This is undesirable from the perspective of generating evidence for type class. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/2256#comment:25> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #10886: Remove the magic from `Any`
by GHC 11 Jul '16

11 Jul '16

#10886: Remove the magic from `Any` -------------------------------------+------------------------------------- Reporter: goldfire | Owner: goldfire Type: task | Status: new Priority: normal | Milestone: 8.0.1 Component: Compiler | Version: 7.10.2 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- The `Any` type used to be very highly magical: a poly-kinded type in a mono-kinded world. Then it became only normally magical: a datatype with a return kind of `k` (instead of `*`). Then it became only a bit magical: a closed type family with no equations. Now that we have non-magical empty closed type families, I believe that we can remove the last bit of magic and just define {{{ type family Any :: k where { } }}} somewhere (`GHC.Types`?), wire it in (like, e.g., `Bool`), and be done with it. Objections? -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10886> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 14

Re: [GHC] #2256: Incompleteness of type inference: must quantify over implication constraints
by GHC 11 Jul '16

11 Jul '16

#2256: Incompleteness of type inference: must quantify over implication constraints -------------------------------------+------------------------------------- Reporter: simonpj | Owner: simonpj Type: bug | Status: new Priority: lowest | Milestone: Component: Compiler (Type | Version: 6.8.2 checker) | Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: Type of failure: GHC rejects | Unknown/Multiple valid program | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Changes (by RyanGlScott): * cc: RyanGlScott (added) -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/2256#comment:24> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0