March 2016 - ghc-tickets

Re: [GHC] #1: Implicit parameters cause strange behavi
by GHC 06 Mar '16

06 Mar '16

#1: Implicit parameters cause strange behavi -----------------------+-------------------- Reporter: nobody | Owner: nobody Type: bug | Status: closed Priority: normal | Milestone: Component: Compiler | Version: 5.0 Resolution: Fixed | Keywords: -----------------------+-------------------- Comment (by Sergei Trofimovich <siarheit@…>): In [changeset:"ade1a461ab4ba3e6de3c4afe9fe9766b7b4e51b3/ghc" ade1a461/ghc]: {{{ #!CommitTicketReference repository="ghc" revision="ade1a461ab4ba3e6de3c4afe9fe9766b7b4e51b3" Fix minimum alignment for StgClosure (Trac #11395) The bug is observed on m68k-linux target as crash in RTS: -- a.hs: main = print 43 $ inplace/bin/ghc-stage1 --make -debug a.hs $ ./a Program terminated with signal SIGSEGV, Segmentation fault. #0 0x80463b0a in LOOKS_LIKE_INFO_PTR_NOT_NULL (p=32858) at includes/rts/storage/ClosureMacros.h:248 (gdb) bt #0 0x80463b0a in LOOKS_LIKE_INFO_PTR_NOT_NULL (p=32858) at includes/rts/storage/ClosureMacros.h:248 #1 0x80463b46 in LOOKS_LIKE_INFO_PTR (p=32858) at includes/rts/storage/ClosureMacros.h:253 #2 0x80463b6c in LOOKS_LIKE_CLOSURE_PTR ( p=0x805aac6e <stg_dummy_ret_closure>) at includes/rts/storage/ClosureMacros.h:258 #3 0x80463e4c in initStorage () at rts/sm/Storage.c:121 #4 0x8043ffb4 in hs_init_ghc (...) at rts/RtsStartup.c:181 #5 0x80455982 in hs_main (...) at rts/RtsMain.c:51 #6 0x80003c1c in main () GHC assumes last 2 pointer bits are tags on 32-bit targets. But here 'stg_dummy_ret_closure' address violates the assumption: LOOKS_LIKE_CLOSURE_PTR (p=0x805aac6e <stg_dummy_ret_closure>) I've added compiler hint for static StgClosure objects to align closures at least by their natural alignment (what GHC assumes). See Note [StgWord alignment]. Signed-off-by: Sergei Trofimovich <siarheit(a)google.com> Test Plan: ran basic test on m68k qemu, it got past ASSERTs Reviewers: simonmar, austin, bgamari Reviewed By: bgamari Subscribers: thomie Differential Revision: https://phabricator.haskell.org/D1974 GHC Trac Issues: #11395 }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/1#comment:2> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #2530: deriving Show adds extra parens for constructor with record syntax
by GHC 06 Mar '16

06 Mar '16

#2530: deriving Show adds extra parens for constructor with record syntax -------------------------------------+------------------------------------- Reporter: spl | Owner: bgamari Type: bug | Status: new Priority: highest | Milestone: 8.0.1 Component: Compiler | Version: 6.8.3 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Phab:D669 Wiki Page: | -------------------------------------+------------------------------------- Comment (by hvr): Replying to [comment:33 RyanGlScott]: > Well, if people find this objectionable, I suppose we also need to review #10104, since that also resulted in a change in `Show` output. There's a big difference though: #2530 wants to fix a deliberate deviation (i.e. many people here consider `{}` binding tighter than function application to be a design-flaw) from the Haskell Report. Those additional brackets are harmless and don't harm the expectation of 'Show' resulting in an expression that parses back to the original. On the other hand, #10104 covers an area outside the Haskell Report (`-XMagicHash`) and more importantly fixes the actual problem that `show (MkT 3#)` results in `MkT 3` which can't be parsed back to the original due to a kind-error. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/2530#comment:35> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #11682: :main doesn't use a bound thread
by GHC 06 Mar '16

06 Mar '16

#11682: :main doesn't use a bound thread -------------------------------------+------------------------------------- Reporter: dmwit | Owner: Type: feature | Status: new request | Priority: normal | 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: -------------------------------------+------------------------------------- Here is a short program: {{{ import Control.Concurrent main = do bound <- isCurrentThreadBound print (bound == rtsSupportsBoundThreads) }}} When compiled (with or without {{{-threaded}}}), it prints {{{True}}}. When executed via {{{runhaskell}}} or via ghci's {{{:main}}} command, it prints {{{False}}}. It would be nicer if it printed {{{True}}} in all of these cases, in particular to ease interactive development of GUI code and the like. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11682> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #11516: Panic, "falls into a hole"
by GHC 06 Mar '16

06 Mar '16

#11516: Panic, "falls into a hole" -------------------------------------+------------------------------------- Reporter: Iceland_jack | Owner: Type: bug | Status: new Priority: normal | 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: -------------------------------------+------------------------------------- {{{#!hs {-# language PolyKinds #-} {-# language FlexibleContexts #-} {-# language ConstraintKinds #-} {-# language FlexibleInstances #-} {-# language FunctionalDependencies #-} import GHC.Types (Constraint) class Ríki (p :: i -> i -> *) class (Ríki p) => Varpi p q f | f -> p q instance Varpi () () f => Varpi (->) (->) (Either f) where }}} causes {{{ % ghci -ignore-dot-ghci /tmp/testing.hs GHCi, version 8.1.20160117: http://www.haskell.org/ghc/ :? for help [1 of 1] Compiling Main ( /tmp/testing.hs, interpreted ) /tmp/testing.hs:11:10: error:ghc: panic! (the 'impossible' happened) (GHC version 8.1.20160117 for x86_64-unknown-linux): fvProv falls into a hole {a13R} Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug > Leaving GHCi. }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11516> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #11662: Regression using NamedFieldPuns with qualified field names
by GHC 06 Mar '16

06 Mar '16

#11662: Regression using NamedFieldPuns with qualified field names -------------------------------------+------------------------------------- Reporter: hesselink | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1-rc2 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: -------------------------------------+------------------------------------- I found a regression where code using the NamedFieldPuns extension and qualified field names doesn't compile with GHC 8 rc2, where it did with GHC 7.10. Minimal reproduction: {{{#!hs module Rec where data Rec = Rec { f :: Integer } -------------------- {-# LANGUAGE NamedFieldPuns #-} module Main where import Rec (Rec (Rec)) import qualified Rec g :: Rec -> Integer g (Rec { Rec.f }) = f }}} On GHC 8 this fails with: {{{ Main.hs:6:10: error: Qualified name in binding position: Rec.f }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11662> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #11524: Something is amiss with quantification in pattern synonym type signatures
by GHC 06 Mar '16

06 Mar '16

#11524: Something is amiss with quantification in pattern synonym type signatures -------------------------------------+------------------------------------- Reporter: bgamari | Owner: Type: bug | Status: new Priority: high | Milestone: 8.0.1 Component: Compiler | Version: 8.0.1-rc1 (Type checker) | 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: -------------------------------------+------------------------------------- Consider this program, {{{#!hs {-# LANGUAGE GADTs #-} {-# LANGUAGE PolyKinds #-} {-# LANGUAGE ExplicitForAll #-} {-# LANGUAGE PatternSynonyms #-} {-# LANGUAGE TypeInType #-} module Test where data AType (a :: k) where AMaybe :: AType Maybe AInt :: AType Int AApp :: forall k1 k2 (a :: k1 -> k2) (b :: k1). AType a -> AType b -> AType (a b) pattern PApp :: () => (fun ~ a b) => AType a -> AType b -> AType fun --pattern PApp :: forall k (fun :: k) k1 (a :: k1 -> k) (b :: k1). -- () => (fun ~ a b) => AType a -> AType b -> AType fun pattern PApp fun arg <- AApp fun arg }}} I would have thought that the two type signatures would be equivalent. However, when I use the quantified signature GHC complains with, {{{ hi.hs:14:34: error: • Expected kind ‘AType a’, but ‘fun’ has kind ‘AType a1’ • In the declaration for pattern synonym ‘PApp’ • Relevant bindings include arg :: AType b1 (bound at hi.hs:14:34) fun :: AType a1 (bound at hi.hs:14:30) }}} Moreover, if I use the un-quantified signature and ask GHCi for the full type signature, it gives me the precisely the quantified type that it rejected previously, {{{ λ> :info PApp pattern PApp :: forall k (fun :: k) k1 (a :: k1 -> k) (b :: k1). () => fun ~ a b => AType a -> AType b -> AType fun -- Defined at hi.hs:14:1 }}} Very odd. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11524> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 13

Re: [GHC] #4029: ghci leaks memory when loading a file
by GHC 06 Mar '16

06 Mar '16

#4029: ghci leaks memory when loading a file -------------------------------------+------------------------------------- Reporter: blarsen | Owner: jme Type: bug | Status: closed Priority: high | Milestone: 8.0.1 Component: GHCi | Version: Resolution: fixed | Keywords: memory leak Operating System: Unknown/Multiple | Architecture: Type of failure: Compile-time | Unknown/Multiple performance bug | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Phab:D1950 Wiki Page: | -------------------------------------+------------------------------------- Changes (by bgamari): * status: patch => closed * resolution: => fixed Comment: Merged to `ghc-8.0` as 6ae616f5be1db6da8bc0c5e36736a76cfea46844. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/4029#comment:22> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #11558: Using Cabal 1.22 against GHC 8.0 results in terrible errors
by GHC 06 Mar '16

06 Mar '16

#11558: Using Cabal 1.22 against GHC 8.0 results in terrible errors -------------------------------------+------------------------------------- Reporter: bgamari | Owner: Type: bug | Status: new Priority: highest | Milestone: 8.0.1 Component: Package | Version: 8.0.1-rc1 system | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Incorrect Unknown/Multiple | warning at compile-time Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- GHC 8.0 will require Cabal >1.22 due to a variety of interface changes (see #10714, Phab:D1780, https://mail.haskell.org/pipermail/ghc- devs/2016-January/010978.html). Unfortunately things currently blow up late in compilation with a seemingly unrelated error when this requirement isn't met. Namely, we complain about missing interface files for dependencies. For instance, {{{ Text/Parsec/Prim.hs:85:1: error: Bad interface file: /root/.cabal/lib/x86_64-linux- ghc-8.0.0.20160204/mtl-2.2.1-9wMhzcfsAXi7cNtl9n7svO/Control/Monad/Cont/Class.hi Something is amiss; requested module mtl-2.2.1@mtl-2.2.1-1af5dabd2d7abb688f2145aec87badc4:Control.Monad.Cont.Class differs from name found in the interface file mtl_9wMhzcfsAXi7cNtl9n7svO:Control.Monad.Cont.Class }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11558> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 12

Re: [GHC] #4029: ghci leaks memory when loading a file
by GHC 06 Mar '16

06 Mar '16

#4029: ghci leaks memory when loading a file -------------------------------------+------------------------------------- Reporter: blarsen | Owner: jme Type: bug | Status: patch Priority: high | Milestone: 8.0.1 Component: GHCi | Version: Resolution: | Keywords: memory leak Operating System: Unknown/Multiple | Architecture: Type of failure: Compile-time | Unknown/Multiple performance bug | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Phab:D1950 Wiki Page: | -------------------------------------+------------------------------------- Comment (by Ben Gamari <ben@…>): In [changeset:"6ca9b15f77e58931953edb7c872b803cb261fce9/ghc" 6ca9b15/ghc]: {{{ #!CommitTicketReference repository="ghc" revision="6ca9b15f77e58931953edb7c872b803cb261fce9" GHCi: Fix load/reload space leaks (#4029) This patch addresses GHCi load/reload space leaks which could be fixed without adversely affecting performance. Test Plan: make test "TEST=T4029" Reviewers: austin, bgamari Reviewed By: bgamari Subscribers: mpickering, thomie Differential Revision: https://phabricator.haskell.org/D1950 GHC Trac Issues: #4029 }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/4029#comment:21> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #11673: Doesn't accept type
by GHC 05 Mar '16

05 Mar '16

#11673: Doesn't accept type -------------------------------------+------------------------------------- Reporter: Iceland_jack | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.1 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: -------------------------------------+------------------------------------- {{{#!hs import Data.Kind import qualified Control.Category as Cat newtype Transformer f g where Transform :: (forall i. f i ~> g i) -> Transformer f g type family (~>) :: k -> k -> Type where (~>) = (->) (~>) = Transformer instance Cat.Category ((~>) :: k -> k -> Type) => Cat.Category (Transformer :: (i -> k) -> (i -> k) -> Type) where id :: forall (f :: i -> k). Transformer f f id = Transform Cat.id }}} works, as well as omitting the instance signature: {{{#!hs id = Transform Cat.id }}} Implicitly quantifying `f` or omitting its kind signature result in the same error: {{{#!hs id :: Transformer f f id = Transform Cat.id -- • Could not deduce (Cat.Category (~>)) -- arising from a use of ‘Cat.id’ -- from the context: Cat.Category (~>) -- bound by the instance declaration at /tmp/tX81.hs:(13,3)-(15,60) -- • In the first argument of ‘Transform’, namely ‘Cat.id’ -- In the expression: Transform Cat.id -- In an equation for ‘id’: id = Transform Cat.id -- Compilation failed. }}} Should it compile -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11673> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1