September 2015 - ghc-tickets

[GHC] #10726: Upgrade MingW-w64 distributions for windows
by GHC 14 Dec '15

14 Dec '15

#10726: Upgrade MingW-w64 distributions for windows -------------------------------------+------------------------------------- Reporter: Phyx- | Owner: Phyx- Type: task | Status: new Priority: normal | Milestone: 7.12.1 Component: Build | Version: 7.11 System | Keywords: | Operating System: Windows Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: #9218 #9014 Differential Revisions: | #10435 -------------------------------------+------------------------------------- This is the second part of splitting up #9218. The goal is to move `GHC x86` from `MingW` to `MingW-w64` and to upgrade both the `x86` and the `x86_64` toolchains shipped with `GHC` to newer versions. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10726> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 67

[GHC] #10250: API Annotations : add Locations in hsSyn were layout occurs
by GHC 14 Dec '15

14 Dec '15

#10250: API Annotations : add Locations in hsSyn were layout occurs -------------------------------------+------------------------------------- Reporter: alanz | Owner: alanz Type: task | Status: new Priority: normal | Milestone: 7.10.2 Component: Compiler | Version: 7.10.1 Keywords: | Operating System: Unknown/Multiple ApiAnnotations | Type of failure: None/Unknown Architecture: | Blocked By: Unknown/Multiple | Related Tickets: Test Case: | Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- At the moment ghc-exactprint, which uses the GHC API Annotations to provide a framework for roundtripping Haskell source code with optional AST edits, has to implement a horrible workaround to manage the points where layout needs to be captured. These are MatchGroup HsDo HsCmdDo HsLet LetStmt HsCmdLet GRHSs To provide a more natural representation, the contents subject to layout rules need to be wrapped in a SrcSpan. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10250> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 14

[GHC] #10238: Update libffi shortly before first GHC 7.12.1 RC
by GHC 14 Dec '15

14 Dec '15

#10238: Update libffi shortly before first GHC 7.12.1 RC -------------------------------------+------------------------------------- Reporter: hvr | Owner: hvr Type: task | Status: new Priority: highest | Milestone: 7.12.1 Component: Runtime | Version: System | Operating System: Unknown/Multiple Keywords: libffi | Type of failure: None/Unknown Architecture: | Blocked By: Unknown/Multiple | Related Tickets: Test Case: | Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- Making this ticket so I don't forget to check for the latest libffi release shortly before the first RC (like I missed that "libffi-3.2.1 was released on November 12, 2014" which would have been in time for 7.10.1RC1) -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10238> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #10845: Incorrect behavior when let binding implicit CallStack object
by GHC 14 Dec '15

14 Dec '15

#10845: Incorrect behavior when let binding implicit CallStack object -------------------------------------+------------------------------------- Reporter: | Owner: nitromaster101 | Type: bug | Status: new 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'm trying out ghc HEAD at 062feee4e7408ad5b9d882e5fed2c700e337db72. {{{#!hs {-# LANGUAGE ImplicitParams #-} import GHC.Types f :: (?loc :: CallStack) => Int f = let y = length $ getCallStack ?loc in if y < 2 then 3 else 4 f2 :: (?loc :: CallStack) => Int f2 = if (length $ getCallStack ?loc) < 2 then 3 else 4 f3 :: (?loc :: [Int]) => Int f3 = let y = length ?loc in if y < 2 then 3 else 4 }}} produces: {{{ [1 of 1] Compiling Main ( Bug.hs, interpreted ) Bug.hs:5:6: warning: Redundant constraint: ?loc::CallStack In the type signature for: f :: (?loc::CallStack) => Int }}} f3 doesn't produce an error which makes me think it's related to CallStack handling. The warning seems accurate: the returned callstack only contains the entry at f. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10845> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 19

[GHC] #10819: Can't generate data decls deriving multiparam type classes with TH
by GHC 14 Dec '15

14 Dec '15

#10819: Can't generate data decls deriving multiparam type classes with TH -------------------------------------+------------------------------------- Reporter: spinda | Owner: Type: feature | Status: new request | Priority: normal | Milestone: Component: Template | Version: 7.10.2 Haskell | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- This applies to both data declarations and (more commonly) newtypes. Example: {{{ {-# LANGUAGE GeneralizedNewtypeDeriving #-} newtype T a = X { unT :: State Int a } deriving (MonadState Int) }}} The newtype declaration is impossible to express in Template Haskell's AST, as the field for the derived classes is {{{[Name]}}} when it should really be {{{Cxt}}}. A workaround is to generate a {{{StandaloneDerivD}}} for each derived multiparam class, but this requires that the user enable the {{{StandaloneDeriving}}} extension which can be unintuitive, especially as there's no way (as far as I can tell) to detect whether this extension is enabled and report a proper error message if it isn't. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10819> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #10741: add flag to dump module and package dependencies
by GHC 14 Dec '15

14 Dec '15

#10741: add flag to dump module and package dependencies -------------------------------------+------------------------------------- Reporter: elaforge | Owner: Type: feature | Status: new request | Priority: normal | Milestone: Component: Compiler | Version: Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- You can already dump module and package dependencies with -M -include-pkg- deps, but it wants to modify a Makefile, so it's not so convenient for things that are not make. For other build systems, it would be convenient to get this on stdout in an easily parseable form. Say X/A.hs imports Y/B.hs, which imports Z/C.hs, and they all use package 'base' and 'containers': {{{ % ghc -Mstdout X/A.hs X/A.h Y/B.hs Y/B.hs Z/C.hs base containers % }}} I don't know about what the output format should be, but assuming no newlines in filenames, and package names never have spaces, it seems simplest to do something like: [src ++ "\n" ++ dest ++ "\n" | (src, dest) <- moduleDeps] ++ "\n" ++ unwords packageDeps It could use haskell syntax like --info, but then you'd need a full haskell parser just to read it, which is not a problem for shake, but probably inconvenient for other build systems. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10741> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

[GHC] #10732: Legal Bang Patterns cannot parse
by GHC 13 Dec '15

13 Dec '15

#10732: Legal Bang Patterns cannot parse -------------------------------------+------------------------------------- Reporter: rem | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.3 (Parser) | Keywords: Bang | Operating System: Unknown/Multiple Pattern Parser | Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- According to the syntax provided by GHC user guide pat ::= !pat Both !!pat and pat : !pat should be accepted, but GHC gives parsing error. For detail see http://stackoverflow.com/questions/31790327/where-can- bangpatterns-appear -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10732> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #10902: Refactor type families in Template Haskell
by GHC 12 Dec '15

12 Dec '15

#10902: Refactor type families in Template Haskell -------------------------------------+------------------------------------- Reporter: goldfire | Owner: Type: task | Status: new Priority: normal | Milestone: 8.0.1 Component: Template | Version: 7.11 Haskell | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- I see that we now have `OpenTypeFamilyD` and `ClosedTypeFamilyD` in Template Haskell. These constructors have quite a lot in common, and I think it's worth having a `TypeFamilyHead` datatype that contains the bits in common. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10902> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 11

[GHC] #10785: ghc-pkg appends colon to db path when piped/redirected
by GHC 12 Dec '15

12 Dec '15

#10785: ghc-pkg appends colon to db path when piped/redirected -------------------------------------+------------------------------------- Reporter: jgertm | Owner: jgertm Type: bug | Status: new Priority: normal | Milestone: Component: ghc-pkg | Version: 7.10.2 Keywords: colon pipe | Operating System: Unknown/Multiple redirect ghc-pkg list | Architecture: | Type of failure: Other Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- Quite simple. Compare the first line of output from {{{#!sh ghc-pkg list ghc-pkg list | head }}} In the second command's output, a colon gets appended at the end of the first line, compared to the first command. This is generally undesired behaviour, since the intent of redirection is to further process the output of the original command. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10785> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #8566: Panic with kindFunResult
by GHC 12 Dec '15

12 Dec '15

#8566: Panic with kindFunResult ------------------------------------+------------------------------------- Reporter: dreixel | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | 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: | ------------------------------------+------------------------------------- The following program: {{{ {-# LANGUAGE DataKinds #-} {-# LANGUAGE PolyKinds #-} {-# LANGUAGE GADTs #-} {-# LANGUAGE FlexibleInstances #-} {-# LANGUAGE ScopedTypeVariables #-} {-# LANGUAGE TypeOperators #-} {-# LANGUAGE UndecidableInstances #-} {-# LANGUAGE MultiParamTypeClasses #-} module Bug where data U (s :: *) = forall k. AA k [U s] data I (u :: U *) (r :: [*]) :: * where A :: I (AA t as) r -- fs unused, but needs to be present for the bug class C (u :: U *) (r :: [*]) (fs :: [*]) where c :: I u r -> I u r instance (C (AA (t (I a ps)) as) ps fs) => C (AA t (a ': as)) ps fs where c A = c undefined }}} crashes a fresh copy of GHC HEAD with the following: {{{ ghc-stage2: panic! (the 'impossible' happened) (GHC version 7.7.20131126 for i386-unknown-linux): kindFunResult k1{tv a24f} [ssk] }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8566> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 21