July 2015 - ghc-tickets

[GHC] #10590: RTS failing with removeThreadFromDeQueue: not found message
by GHC 23 Sep '15

23 Sep '15

#10590: RTS failing with removeThreadFromDeQueue: not found message -------------------------------------+------------------------------------- Reporter: qnikst | 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: | -------------------------------------+------------------------------------- Under certain circumstances I'm facing RTS error, this happens when one thread is reading socket and another one closes it. Seems like that it happens when smth else is involved as I failed to trim this example Error that I see: {{{ % ./test nid://127.0.0.1:8080:0 1 test: internal error: removeThreadFromDeQueue: not found (GHC version 7.10.1 for x86_64_unknown_linux) Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug zsh: abort ./test }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10590> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 8

[GHC] #10660: .dyn_o isn't generated for .hsig files with -dynamic-too
by GHC 23 Sep '15

23 Sep '15

#10660: .dyn_o isn't generated for .hsig files with -dynamic-too -------------------------------------+------------------------------------- Reporter: spinda | 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 | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- This causes the linker to error out later on, as it expects a .dyn_o file to exist for each .hsig file. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10660> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #10689: compiling singletons-1.1.2.1 as -O1 -fspec-constr fails as 'Template variable unbound in rewrite rule'
by GHC 23 Sep '15

23 Sep '15

#10689: compiling singletons-1.1.2.1 as -O1 -fspec-constr fails as 'Template variable unbound in rewrite rule' -------------------------------------+------------------------------------- Reporter: slyfox | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.2-rc2 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- First observed on singletons-1.1.2.1 package ghc-7.10.2. I've tried to distill it down to a single file. The file manages to crash both GHC and GHCi. GHCi: {{{ $ inplace/bin/ghc-stage2 -hide-all-packages -package=base --interactive List.hs GHCi, version 7.11.20150723: http://www.haskell.org/ghc/ :? for help [1 of 1] Compiling List ( List.hs, interpreted ) ghc-stage2: panic! (the 'impossible' happened) (GHC version 7.11.20150723 for x86_64-unknown-linux): floatExpr tick break<2>() Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug }}} GHC: {{{ $ inplace/bin/ghc-stage2 -hide-all-packages -package=base --make List.hs -dsuppress-all -dsuppress-uniques [1 of 1] Compiling List ( List.hs, List.o ) ghc-stage2: panic! (the 'impossible' happened) (GHC version 7.11.20150723 for x86_64-unknown-linux): Template variable unbound in rewrite rule m [a_afdP, x, m, ipv, ipv, sc, sc, sc, sg] [a_afdP, x, m, ipv, ipv, sc, sc, sc, sg] [TYPE a_afdP, TYPE x, TYPE Let1627448493XsSym4 x m ipv ipv, sc, (SCons @~ ... sc sc) `cast` ...] [TYPE a_afdP, TYPE x, TYPE Let1627448493XsSym4 x ipv ipv ipv, sc, (SCons @~ ... ipv ipv) `cast` ...] Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug }}} The "minimal" single-file sample is attached. If you need original selfcontained reproducer (79 modules, template haskell code) this is the version i've started with: http://code.haskell.org/~slyfox/unbound-template-var.tar.gz Adjusting path to GHC in '''./trigger-a-bug.bash''' should be enough. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10689> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 9

[GHC] #10413: Incorrect offsets for array size indexing
by GHC 23 Sep '15

23 Sep '15

#10413: Incorrect offsets for array size indexing -------------------------------------+------------------------------------- Reporter: fryguybob | Owner: Type: task | Status: new Priority: low | Milestone: Component: Compiler | Version: 7.10.1 (CodeGen) | Operating System: Unknown/Multiple Keywords: | Type of failure: Other Architecture: | Blocked By: Unknown/Multiple | Related Tickets: Test Case: | Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- There are several cases in the array code generation where a byte offset is used when a word offset is needed. For example ([https://git.haskell.org/ghc.git/blob/eb6ca851f553262efe0824b8dcbe64952de496… context]): {{{#!hs emitPrimOp dflags [res] SizeofSmallArrayOp [arg] = emit $ mkAssign (CmmLocal res) (cmmLoadIndexW dflags arg (fixedHdrSizeW dflags + oFFSET_StgSmallMutArrPtrs_ptrs dflags) (bWord dflags)) }}} This only works because `oFFSET_StgSmallMutArrPtrs_ptrs dflags` is zero. I don't have time to fix this now, but I'm reporting it so it isn't forgotten. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10413> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #10715: Possible regression in Coercible a (X a) between 7.8 and 7.10
by GHC 23 Sep '15

23 Sep '15

#10715: Possible regression in Coercible a (X a) between 7.8 and 7.10 -------------------------------------+------------------------------------- Reporter: inaki | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: GHC rejects Unknown/Multiple | valid program Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- In upgrading to7.10, code of the form {{{#!hs {-# LANGUAGE FlexibleContexts #-} import Data.Coerce (coerce, Coercible) data X a doCoerce :: Coercible a (X a) => a -> X a doCoerce = coerce }}} fails to compile in 7.10.1 and 7.10.2 with the error {{{ testCoerce.hs:6:13: Could not deduce (a ~ X a) from the context (Coercible a (X a)) bound by the type signature for doCoerce :: Coercible a (X a) => a -> X a at testCoerce.hs:6:13-41 ‘a’ is a rigid type variable bound by the type signature for doCoerce :: Coercible a (X a) => a -> X a at testCoerce.hs:6:13 Relevant role signatures: type role X phantom In the ambiguity check for the type signature for ‘doCoerce’: doCoerce :: forall a. Coercible a (X a) => a -> X a To defer the ambiguity check to use sites, enable AllowAmbiguousTypes In the type signature for ‘doCoerce’: doCoerce :: Coercible a (X a) => a -> X a }}} while it works in 7.8.4. Surprisingly (to me at least), the code works in 7.10.1 and 7.10.2 if I change it to {{{#!hs {-# LANGUAGE FlexibleContexts #-} import Data.Coerce (coerce, Coercible) data X a doCoerce :: Coercible a (X b) => a -> X a doCoerce = coerce }}} while it fails to compile in 7.8.4 with the error {{{ testCoerce.hs:6:13: Could not coerce from ‘a’ to ‘X b0’ because ‘a’ and ‘X b0’ are different types. arising from the ambiguity check for ‘doCoerce’ from the context (Coercible a (X b)) bound by the type signature for doCoerce :: Coercible a (X b) => a -> X a at testCoerce.hs:6:13-41 The type variable ‘b0’ is ambiguous In the ambiguity check for: forall a b. Coercible a (X b) => a -> X a To defer the ambiguity check to use sites, enable AllowAmbiguousTypes In the type signature for ‘doCoerce’: doCoerce :: Coercible a (X b) => a -> X a }}} The coercion pattern may look a bit funny, but it is rather useful when one has newtypes of the form {{{#!hs newtype Y = Y (ForeignPtr Y) }}} which appear naturally when writing bindings to C libraries, and one wants to get access to the underlying ForeignPtr from Y (here X -> ForeignPtr). The relevant Coercible instance here is Coercible Y (ForeignPtr Y), as above. I would have expected the version with context "Coercible a (X a)" to be accepted, as 7.8.4 does, since it seems to be a specialization of the more general coerce, but maybe I am missing something? -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10715> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 18

[GHC] #10713: Type family not reducing over data family
by GHC 23 Sep '15

23 Sep '15

#10713: Type family not reducing over data family -------------------------------------+------------------------------------- Reporter: acowley | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.2 (Type checker) | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- Given this code, {{{ type family TEq t s where TEq t t = 'True TEq t s = 'False data family T a }}} I expected this GHCi interaction to reduce: `:kind! TEq (T Int) (T Bool)` But it does not. It does reduce (to `'True`) if you instead ask, `:kind! TEq (T Int) (T Int)` Tested on GHC 7.10.2 -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10713> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #10665: INLINE breaks rewrite rules when '-g' is used
by GHC 23 Sep '15

23 Sep '15

#10665: INLINE breaks rewrite rules when '-g' is used -------------------------------------+------------------------------------- Reporter: slyfox | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.2-rc2 (CodeGen) | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Compile-time Unknown/Multiple | crash Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- The bug is found when building conduit-1.2.4.2 package with '-O2 -g' options. The distilled sample looks like that: {{{#!hs {-# LANGUAGE BangPatterns #-} module RewriteBug (bug) where bug :: () -> () bug () = bug () {-# NOINLINE bug #-} a2 :: () a2 = () {-# INLINE[1] a2 #-} {-# RULES "bug a2" [0] bug a2 = () #-} {- Crashes as: $ inplace/bin/ghc-stage2 -c -O1 -fforce-recomp RewriteBug.hs -g ghc-stage2: panic! (the 'impossible' happened) (GHC version 7.11.20150721 for x86_64-unknown-linux): Tick in rule () -} }}} My theory of sequence of actions is the following: - rewrite rule gets read as-is by GHC (gentle phase) - a2 INLINE changes LHS of rewrite rule (phase 1) - when time comes to apply 'bug a2' rule GHC detects INLINE problem (phase 0) In real code it happened across multiple files. The bug is reproducible in both ghc-7.10.2-rc2 and today's HEAD. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10665> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #10668: Missing brackets in import hint with TypeOperators
by GHC 23 Sep '15

23 Sep '15

#10668: Missing brackets in import hint with TypeOperators -------------------------------------+------------------------------------- Reporter: bjmprice | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Incorrect Unknown/Multiple | warning at compile-time Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- {{{#!hs import Data.Type.Equality(Refl) }}} errors with {{{ error: In module ‘Data.Type.Equality’: ‘Refl’ is a data constructor of ‘:~:’ To import it use ‘import’ Data.Type.Equality( :~:( Refl ) ) or ‘import’ Data.Type.Equality( :~:(..) ) }}} But the code there does not parse, one needs {{{#!hs import Data.Type.Equality((:~:)(Refl)) }}} instead (note the extra brackets!). -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10668> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #10495: Total inability to infer type for coerce
by GHC 23 Sep '15

23 Sep '15

#10495: Total inability to infer type for coerce -------------------------------------+------------------------------------- Reporter: goldfire | Owner: goldfire 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 | Blocked By: Test Case: | Related Tickets: Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- If I say {{{ import Data.Coerce foo = coerce }}} I get {{{ Couldn't match representation of type ‘a0’ with that of ‘b0’ ‘a0’ is untouchable inside the constraints: () bound by the inferred type of foo :: a0 -> b0 at /Users/rae/temp/bug/App.hs:3:1-12 Relevant bindings include foo :: a0 -> b0 (bound at /Users/rae/temp/bug/App.hs:3:1) In the expression: coerce In an equation for ‘foo’: foo = coerce }}} How awful. Eta-expanding `foo` works fine, though. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10495> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #9970: Export more types in GHC.RTS.Flags
by GHC 23 Sep '15

23 Sep '15

#9970: Export more types in GHC.RTS.Flags -------------------------------------+------------------------------------- Reporter: | Owner: RyanGlScott | Status: new Type: feature | Milestone: request | Version: 7.10.1-rc1 Priority: normal | Operating System: Unknown/Multiple Component: | Type of failure: Other libraries/base | Blocked By: Keywords: | Related Tickets: Architecture: | Unknown/Multiple | Test Case: | Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- There are some ADTs in {{{GHC.RTS.Flags}}} that have arguments whose types are not exported. These types are: * {{{GiveGCStats}}} (found in {{{GCFlags}}}) * {{{DoCostCentres}}} (found in {{{CCFlags}}}) * {{{DoHeapProfile}}} (found in {{{ProfFlags}}}) * {{{DoTrace}}} (found in {{{TraceFlags}}}) This makes it pretty difficult to do anything useful with them, since they can't be pattern-matched. In addition, there are some unexported type synonyms: * {{{type Time = Word64}}} * {{{type Nat = #{type unsigned int} }}} (likely a {{{Word32}}}) The Haddock docs for {{{GHC.RTS.Flags}}} don't give much of an indication of what {{{Time}}} and {{{Nat}}} are, and since they're already documented in [https://phabricator.haskell.org/diffusion/GHC/browse/master/libraries/base/… the source], they should probably be exported. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9970> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7