December 2015 - ghc-tickets

[GHC] #11135: Migrate more of Data.Functor.* from transformers to base.
by GHC 01 Jan '16

01 Jan '16

#11135: Migrate more of Data.Functor.* from transformers to base. -------------------------------------+------------------------------------- Reporter: hvr | Owner: Type: task | Status: new Priority: highest | Milestone: 8.0.1 Component: | Version: 7.10.2 libraries/base | 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: -------------------------------------+------------------------------------- This was proposed in https://mail.haskell.org/pipermail/libraries/2015-July/026014.html Edward says the following modules are up for migration: - `Data.Functor.Classes` - `Data.Functor.Product` - `Data.Functor.Sum` - `Data.Functor.Compose` related/optionally/undecided: - `Data.Functor.Reverse` - `Control.Applicative.Backwards` not moved because redundant (c.f. `Const`): - `Data.Functor.Constant` -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11135> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 27

[GHC] #11296: T8726 fails on arm
by GHC 01 Jan '16

01 Jan '16

#11296: T8726 fails on arm -------------------------------------+------------------------------------- Reporter: bgamari | Owner: bgamari Type: bug | Status: new Priority: normal | Milestone: 8.0.1 Component: Compiler | Version: 7.10.3 Keywords: | Operating System: Linux Architecture: arm | Type of failure: Incorrect result | at runtime Test Case: T8726 | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- {{{ cd ./numeric/should_run && ./T8726 </dev/null > T8726.run.stdout 2> T8726.run.stderr Wrong exit code (expected 0 , actual 1 ) Stdout: Stderr: T8726: user error (divMod0 -2 79228162514264337593543950336) }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11296> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #11320: Various API Annotations fixes
by GHC 01 Jan '16

01 Jan '16

#11320: Various API Annotations fixes -------------------------------------+------------------------------------- Reporter: alanz | Owner: Type: task | Status: new Priority: normal | Milestone: 8.0.1 Component: Compiler | Version: 8.1 Keywords: | Operating System: Unknown/Multiple ApiAnnotations | Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- - Export unicodeAnn from GHC - unicodeAnn for Annlarrowtail was wrong - Use actual source for a CImport SourceText GHC master commit: https://phabricator.haskell.org/rGHC25e4556d97429e95ddb5972f6e7e6599ef902e9c -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11320> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #11291: DfltProb1(optasm): panic CoreToStg.myCollectArgs
by GHC 01 Jan '16

01 Jan '16

#11291: DfltProb1(optasm): panic CoreToStg.myCollectArgs -------------------------------------+------------------------------------- Reporter: thomie | Owner: Type: bug | Status: new Priority: highest | Milestone: 8.0.1 Component: Compiler | Version: 7.11 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: typecheck/should_compile/DfltProb1 | (optasm) | Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- The program looks like this: {{{ module DfltProb1 where import Control.Monad.ST import Prelude hiding (traverse) traverse :: a -> ST s [a] traverse = undefined -- WORKS with signature test :: Num a => [a] test = runST (traverse 1) main = print test }}} This is the failure: {{{ $ ghc-7.11.20151225 -O DfltProb1.hs ghc-stage2: panic! (the 'impossible' happened) (GHC version 7.11.20151224 for x86_64-unknown-linux): CoreToStg.myCollectArgs (case traverse of _ [Occ=Dead] { }) realWorld# }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11291> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 10

[GHC] #11322: Dysfunctional `__GLASGOW_HASKELL_TH` macro
by GHC 31 Dec '15

31 Dec '15

#11322: Dysfunctional `__GLASGOW_HASKELL_TH` macro -------------------------------------+------------------------------------- Reporter: hvr | Owner: Type: bug | Status: new Priority: highest | Milestone: 8.0.1 Component: Compiler | Version: 7.10.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): Phab:#1723 | Wiki Page: -------------------------------------+------------------------------------- TODO -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11322> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #8182: Parser.y.pp needs special treatment with -fcmm-sink
by GHC 31 Dec '15

31 Dec '15

#8182: Parser.y.pp needs special treatment with -fcmm-sink ------------------------------------+------------------------------------- Reporter: thoughtpolice | Owner: thoughtpolice Type: bug | Status: new Priority: highest | Milestone: 7.8.1 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: | ------------------------------------+------------------------------------- This bug is really more of a reminder, but it's critical for the 7.8.1 release. Right now, `Parser.y.pp` uses a very weird hack to pass `-fcmm-sink` to the compiler. Here's the relevant comment: {{{ -- CPP tricks because we want the directives in the output of the -- first CPP pass. -- -- Clang note, 6/17/2013 by aseipp: It is *extremely* important (for -- some reason) that there be a line of whitespace between the two -- definitions here, and the subsequent use of __IF_GHC_77__ - this -- seems to be a bug in clang or something, where having the line of -- whitespace will make the preprocessor correctly format the rendered -- lines in the 'two step' CPP pass. No, this is not a joke. #define __IF_GHC_77__ #if __GLASGOW_HASKELL__ >= 707 #define __ENDIF__ #endif __IF_GHC_77__ -- Required on x86 to avoid the register allocator running out of -- stack slots when compiling this module with -fPIC -dynamic. {-# OPTIONS_GHC -fcmm-sink #-} __ENDIF__ }}} This is because the parser is first preprocessed before being run with GHC (which again preprocesses,) so we want the resulting `#ifdef` in the final `.hs` file. Things to note: * We really shouldn't be doing this, it's amazingly fragile. I think the correct thing to do is to ensure the build system correctly passes `-fcmm- sink` during the stage[2,3] build. * `./configure.ac` needs to check to see if the bootstrapping compiler is `ghc >= 7.7` (or `7.8`) and if it is, ''also'' pass `-fcmm-sink` during the stage1 build. * This hack needs to be removed once we can rely on >= 7.8 for bootstrap. Probably something like the hypothetical 7.12-7.14 timeframe. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8182> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #9734: Provide macro __GLASGOW_HASKELL_TH__
by GHC 31 Dec '15

31 Dec '15

#9734: Provide macro __GLASGOW_HASKELL_TH__ -------------------------------------+------------------------------------- Reporter: nomeata | Owner: Type: feature request | Status: new Priority: normal | Milestone: 7.10.1 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: -------------------------------------+------------------------------------- The Debian packaging added a few hacks to create a CPP macro that can be used to distinguish compilers with Template Haskell support from those that don’t have that support. We need this to compiler some code on non- TH architectures. It works, but we found that we cannot upstream such patches, which is a shame. I propose that among the GHC-defined macros listed at https://www.haskell.org/ghc/docs/latest/html/users_guide/options- phases.html#c-pre-processor there is also `__GLASGOW_HASKELL_TH__=YES`, which is available if the compiler supports splices, or ``__GLASGOW_HASKELL_TH__=NO` if not. (I propose a two-valued comparison so that library authors can decide themselves what the default should be if this symbol is not defined, some might want to use TH on older compilers, some not.) I hope this can be part of 7.10. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9734> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #11324: Missing Kind Inference
by GHC 31 Dec '15

31 Dec '15

#11324: Missing Kind Inference -------------------------------------+------------------------------------- Reporter: crockeea | Owner: Type: bug | Status: new Priority: normal | Milestone: 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 Rev(s): | Wiki Page: -------------------------------------+------------------------------------- The following example doesn't compile, but I expected it to. {{{ {-# LANGUAGE RankNTypes, ConstraintKinds, ScopedTypeVariables, KindSignatures, PolyKinds, DataKinds, FlexibleInstances, UndecidableInstances, TypeFamilies #-} module Test where data Proxy a data Tagged t s = Tag s type family CharOf fp :: k class Reflects (a :: k) where value :: Proxy a instance Reflects (a :: Bool) type MyConstraint (x :: Bool) = (x~x) foo :: forall fp . (MyConstraint (CharOf fp)) => Tagged fp Int foo= let x = value::Proxy (CharOf fp) in Tag 2 }}} The error in 7.10.2 (unable to test with HEAD) is `Could not deduce (Reflects k (CharOf k fp)) arising from a use of ‘value’`, basically that it was unable to figure out the kind of `CharOf fp`. I think GHC should know the kind from the constraint on `foo`. I've found two workarounds: {{{ foo :: forall fp . (MyConstraint (CharOf fp)) => Tagged fp Int foo= let x = value::Proxy (CharOf fp :: Bin) in Tag 2 }}} and {{{ foo :: forall fp x . (MyConstraint x, x~CharOf fp) => Tagged fp Int foo= let x = value::Proxy x in Tag 2 }}} but both seem unnecessary. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11324> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #11122: Ambiguous inferred type causes a panic
by GHC 31 Dec '15

31 Dec '15

#11122: Ambiguous inferred type causes a panic -------------------------------------+------------------------------------- Reporter: tuplanolla | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.2 (Type checker) | Keywords: | Operating System: Linux Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Consider the following program. {{{#!hs {-# LANGUAGE PartialTypeSignatures #-} module Main where import Text.Parsec import Text.Parsec.String parser :: Parser _ parser = read <$> many digit data Wrapper = Wrapper Int wrapperParser = Wrapper <$> parser }}} I am not sure whether it is valid or not, but it breaks the type checker as follows. {{{#!hs Program.hs:9:1: No instance for (Read w_) When checking that ‘parser’ has the specified type parser :: forall w_. Parser w_ Probable cause: the inferred type is ambiguous Program.hs:13:29: Couldn't match type ‘w_’ with ‘Int’ ‘w_’ is untouchable inside the constraints () bound by the inferred type of wrapperParser :: ParsecT String () Data.Functor.Identity.Identity Wrapper at Program.hs:13:1-34ghc: panic! (the 'impossible' happened) (GHC version 7.10.2 for x86_64-unknown-linux): No skolem info: w__avnz[sk] }}} Disabling `PartialTypeSignatures` makes no difference. Switching from `parsec-3.1.9` to `megaparsec-4.1.1` and changing `many digit` to `some digitChar` does not have an effect either. Replacing `_` with `Int` seems to be the only obvious way to make the problem disappear. Alas that defeats the point of using partial type signatures or typed holes in the first place. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11122> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 9

[GHC] #11245: Non-exhaustive pattern, "Patterns not matched" list is empty
by GHC 31 Dec '15

31 Dec '15

#11245: Non-exhaustive pattern, "Patterns not matched" list is empty -------------------------------------+------------------------------------- Reporter: osa1 | Owner: 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 Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Example: {{{#!haskell module Main where maybeOdd :: Int -> Maybe Int maybeOdd i = if odd i then Just i else Nothing main :: IO () main = do let x = maybeOdd 10 let a | Just i <- x , odd i = True | Nothing <- x = False print x print a }}} Warning printed by GHC HEAD: {{{ Exhaustive.hs:10:7: warning: Pattern match(es) are non-exhaustive In an equation for ‘a’: Patterns not matched: Linking Exhaustive ... }}} The problem with this message is; if it couldn't come up with an example unmatched pattern, then how can it know that the pattern is non- exhaustive? If it came up with an example, why is that example not printed? UPDATE: I just realized it's actually worse that I first thought. If I change {{{a}}} in this example: {{{#!haskell let a | Just i <- x = True }}} This message is printed: {{{ [1 of 1] Compiling Main ( Exhaustive.hs, Exhaustive.o ) Exhaustive.hs:10:7: warning: Pattern match(es) are non-exhaustive In an equation for ‘a’: Patterns not matched: Exhaustive.hs:10:16: warning: Defined but not used: ‘i’ Linking Exhaustive ... }}} NOTE: Tried with GHC 7.10 too. It seems like in the case where the checks are not exhaustive, both 7.10 and HEAD are giving the same warning(with empty list of non-checked patterns). HEAD is better in detecting exhaustive patterns. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11245> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 12