- ghc-tickets - Haskell.org

[GHC] #13987: T13701 fails sporadically
by GHC 20 Jul '17

20 Jul '17

#13987: T13701 fails sporadically -------------------------------------+------------------------------------- Reporter: duog | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Test Suite | Version: 8.3 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Other Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- T13701 fails sporadically, comments in Phab:D3586 indicate that it happens while the system is under load. I reproduced this issue by duplicating T13701 7 times, making tests T13701[A-H], then running {{{ make test TEST="T13701A T13701B T13701C T13701D T13701E T13701F T13701G T13701H THREADS=8" }}} on my box with 4 cpus. Comparing profiling reports of the test under and not under load, I found the additional allocations to be in SysTools.builderMainLoop.loop. This function busy-waits from when 2 EOFs are received until getProcessExitCode returns a Just. That this is the cause was verified by changing {{{ | otherwise -> loop chan hProcess t p exitcode }}} to {{{ | otherwise -> threadDelay 10000 >> loop chan hProcess t p exitcode }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13987> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #13983: Type synonym referred to as datatype in error message
by GHC 20 Jul '17

20 Jul '17

#13983: Type synonym referred to as datatype in error message -------------------------------------+------------------------------------- Reporter: RyanGlScott | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 (Type checker) | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Poor/confusing Unknown/Multiple | error message Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- To reproduce the issue, compile this file with GHC 8.0.1, 8.0.2, 8.2.1, or HEAD: {{{#!hs {-# LANGUAGE PolyKinds #-} {-# LANGUAGE RankNTypes #-} module Bug where import Data.Proxy type Wat = forall (a :: k). Proxy a }}} {{{ GHCi, version 8.3.20170706: http://www.haskell.org/ghc/ :? for help Loaded GHCi configuration from /home/rgscott/.ghci [1 of 1] Compiling Bug ( Bug.hs, interpreted ) Bug.hs:7:1: error: • Kind variable ‘k’ is implicitly bound in datatype ‘Wat’, but does not appear as the kind of any of its type variables. Perhaps you meant to bind it (with TypeInType) explicitly somewhere? Type variables with inferred kinds: (k :: *) • In the type declaration for ‘Wat’ | 7 | type Wat = forall (a :: k). Proxy a | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ }}} The error message mistakenly refers to `Wat` as a datatype. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13983> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #13680: Can't use TypeApplications with empty list expression
by GHC 20 Jul '17

20 Jul '17

#13680: Can't use TypeApplications with empty list expression -------------------------------------+------------------------------------- Reporter: RyanGlScott | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 (Type checker) | Keywords: | Operating System: Unknown/Multiple TypeApplications | Architecture: | Type of failure: GHC rejects Unknown/Multiple | valid program Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Why can't I do this? {{{#!hs {-# LANGUAGE TypeApplications #-} module Bug where foo :: [Int] foo = [] @Int }}} Compiling this with GHC 8.0.1, 8.0.2, 8.2.1, or HEAD gives me: {{{ $ /opt/ghc/head/bin/ghci Bug2.hs GHCi, version 8.3.20170509: http://www.haskell.org/ghc/ :? for help Loaded GHCi configuration from /home/rgscott/.ghci [1 of 1] Compiling Bug ( Bug2.hs, interpreted ) Bug2.hs:5:7: error: • Cannot apply expression of type ‘[a0]’ to a visible type argument ‘Int’ • In the expression: [] @Int In an equation for ‘foo’: foo = [] @Int | 5 | foo = [] @Int | ^^^^^^^ }}} This seems really strange, since I can use `TypeApplications` with expressions like `Nothing @Int` without issues. According to GHCi: {{{ $ /opt/ghc/head/bin/ghci GHCi, version 8.3.20170509: http://www.haskell.org/ghc/ :? for help Loaded GHCi configuration from /home/rgscott/.ghci λ> :set -XTypeApplications -fprint-explicit-foralls λ> :type +v [] [] :: forall {a}. [a] λ> :type +v Nothing Nothing :: forall a. Maybe a }}} The type variable for `[]` isn't visible! But I don't see any reason why it shouldn't be, especially since, conceptually, the data type declaration for lists is: {{{#!hs data [] a = [] | a : [a] }}} I suspect that `[]`'s tyvar binder visibility is not wired into GHC correctly. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13680> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 15

[GHC] #13434: hs_try_putmvar003 is timing out / segfaulting
by GHC 20 Jul '17

20 Jul '17

#13434: hs_try_putmvar003 is timing out / segfaulting -------------------------------------+------------------------------------- Reporter: ezyang | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Runtime | Version: 8.1 System | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Runtime crash Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- This is basically reproduceable every time I validate. During validation, the program times out, but when I run it by hand, it segfaults. Running in debug sanity mode, I get: {{{ hstry: internal error: MVAR_CLEAN on mutable list (GHC version 8.3.20170316 for x86_64_unknown_linux) Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug }}} Simon Marlow, I'm happy to investigate more if you can't repro. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13434> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 11

[GHC] #13915: GHC 8.2 regression: "Can't find interface-file declaration" for promoted data family instance
by GHC 20 Jul '17

20 Jul '17

#13915: GHC 8.2 regression: "Can't find interface-file declaration" for promoted data family instance -------------------------------------+------------------------------------- Reporter: RyanGlScott | Owner: (none) Type: bug | Status: new Priority: high | Milestone: 8.2.1 Component: Compiler | Version: 8.2.1-rc2 Keywords: TypeInType, | Operating System: Unknown/Multiple TypeFamilies | Architecture: | Type of failure: GHC rejects Unknown/Multiple | valid program Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Due to #12088, you can't define a data family instance and promote it in the same module. One could, up until GHC 8.2, work around this using (somewhat obscure) wisdom: define the data family instance in a separate module from where it's promoted. For example, `Bug` typechecks in GHC 8.0.1 and 8.0.2: {{{#!hs {-# LANGUAGE TypeFamilies #-} module Foo where data family T a data instance T Int = MkT }}} {{{#!hs {-# LANGUAGE TypeInType #-} module Bug where import Foo data Proxy (a :: k) data S = MkS (Proxy 'MkT) }}} However, this stopped typechecking in GHC 8.2: {{{ $ /opt/ghc/8.2.1/bin/ghci Bug.hs GHCi, version 8.2.0.20170623: http://www.haskell.org/ghc/ :? for help Loaded GHCi configuration from /home/rgscott/.ghci [1 of 2] Compiling Foo ( Foo.hs, interpreted ) [2 of 2] Compiling Bug ( Bug.hs, interpreted ) Bug.hs:1:1: error: Can't find interface-file declaration for variable Foo.$tc'MkT Probable cause: bug in .hi-boot file, or inconsistent .hi file Use -ddump-if-trace to get an idea of which file caused the error | 1 | {-# LANGUAGE TypeInType #-} | ^ }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13915> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #13948: GHC incorrectly suggests a data constructor is in scope in 8.2
by GHC 19 Jul '17

19 Jul '17

#13948: GHC incorrectly suggests a data constructor is in scope in 8.2 -------------------------------------+------------------------------------- Reporter: RyanGlScott | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.2.1-rc2 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Poor/confusing Unknown/Multiple | error message Test Case: | Blocked By: Blocking: | Related Tickets: #13568 Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- After commit 343cb32d0983f576d344a2d04a35c3fd6eecf2c5 (the fix for #13568), which debuted in GHC 8.2, GHC thinks that out-of-scope identifiers are actually in-scope! {{{#!hs module Bug where f :: T f = undefined }}} In GHC 8.2.1-rc3, the error message is: {{{ $ /opt/ghc/8.2.1/bin/ghc Bug.hs [1 of 1] Compiling Bug ( Bug.hs, Bug.o ) Bug.hs:3:6: error: Not in scope: type constructor or class ‘T’ A data constructor of that name is in scope; did you mean DataKinds? | 3 | f :: T | ^ }}} Compare this to GHC 8.0.2, which gives a more sensible error message: {{{ $ /opt/ghc/8.0.2/bin/ghc Bug.hs [1 of 1] Compiling Bug ( Bug.hs, Bug.o ) Bug.hs:3:6: error: Not in scope: type constructor or class ‘T’ }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13948> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #13899: Improve [-Wmissing-home-modules] error message
by GHC 19 Jul '17

19 Jul '17

#13899: Improve [-Wmissing-home-modules] error message -------------------------------------+------------------------------------- Reporter: alanz | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: 8.2.1 Component: Compiler | Version: 8.2.1-rc2 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: -------------------------------------+------------------------------------- At the moment when this warning is triggered, GHC emits something like {{{#!hs <no location info>: warning: [-Wmissing-home-modules] Modules are not listed in command line: Language.Haskell.GHC.ExactPrint }}} It is not clear that this is actually reporting a problem in the cabal file which has incomplete "other-modules" for one or other target. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13899> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #13996: Non-cheap primop is duplicated
by GHC 19 Jul '17

19 Jul '17

#13996: Non-cheap primop is duplicated -------------------------------------+------------------------------------- Reporter: simonpj | Owner: (none) 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: -------------------------------------+------------------------------------- Conal Elliot [https://mail.haskell.org/pipermail/ghc- devs/2017-July/014407.html writes]: I have {{{ exampleC :: Double -> Double -> Double exampleC = \ t -> let s = sin t in \ x -> x + s }}} But the `sin` ends up being pushed inside the lambda by the optimiser, like so: {{{ exampleC = \ t x -> x + sin t }}} So if I do `map (exampleC v) xs` I end up computing `sin v` once for each element of `xs` rather than once. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13996> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #11095: -O0 -g slows GHC down on list literals (compared to -O0 without -g)
by GHC 19 Jul '17

19 Jul '17

#11095: -O0 -g slows GHC down on list literals (compared to -O0 without -g) -------------------------------------+------------------------------------- Reporter: slyfox | 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: -------------------------------------+------------------------------------- Tried to buld syb-0.6 testsuite with '-O0 -g' flags today and noticed compiler hung for a couple of hours. Distilled example from syb-0.6 testsuite (exact example attached): {{{#!hs {-# LANGUAGE NoImplicitPrelude #-} {-# OPTIONS_GHC -O0 #-} module Bug (bug) where {-# NOINLINE f #-} f :: a -> a f v = v bug :: [()] bug = f v v :: [()] v = [(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),(),() ,(),(),(),(),() ... --- a lot of these ] }}} {{{ $ time ghc -c -fforce-recomp -O0 Bug.hs real 0m0.333s user 0m0.288s sys 0m0.042s $ time ghc -c -fforce-recomp -O0 -g Bug.hs real 0m40.241s user 0m40.021s sys 0m0.070s }}} For some reason '''-g''' manages slow things down at a large factor. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11095> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 20

[GHC] #13997: GHCi panic: "Loading temp shared object failed" when adding a new module
by GHC 19 Jul '17

19 Jul '17

#13997: GHCi panic: "Loading temp shared object failed" when adding a new module -------------------------------------+------------------------------------- Reporter: mbieleck | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.2.1-rc3 Keywords: panic GHCi | Operating System: Unknown/Multiple Architecture: x86_64 | Type of failure: Compile-time (amd64) | crash or panic Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Files: {{{#!hs -- Bug.hs module Bug where import Bug2 bug = bug2 -- Bug2.hs module Bug2 where bug2 = () }}} Steps to reproduce: 1. Create the files as above 2. Start GHCi: {{{ $ ghc-8.2.0.20170704 --interactive -fobject-code λ> :load Bug Ok, 2 modules loaded. }}} 3. Without exiting GHCi, add a new file: {{{#!hs -- NewModule.hs module NewModule where newFn = () }}} and change `Bug2.hs` to: {{{#!hs module Bug2 where import NewModule bug2 = newFn }}} to introduce a transitive dependency `Bug -> Bug2 -> NewModule`. 4. In GHCi {{{ λ> :reload [2 of 3] Compiling Bug2 ( Bug2.hs, Bug2.o ) Ok, 3 modules loaded. λ> bug ghc: panic! (the 'impossible' happened) (GHC version 8.2.0.20170704 for x86_64-unknown-linux): Loading temp shared object failed: /tmp/ghc22212_0/libghc_4.so: undefined symbol: NewModule_newFn_closure Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug }}} The problem also exists without `-fobject-code`. The error message is different: {{{ λ> bug ghc: ^^ Could not load 'NewModule_newFn_closure', dependency unresolved. See top entry above. ByteCodeLink.lookupCE During interactive linking, GHCi couldn't find the following symbol: NewModule_newFn_closure This may be due to you not asking GHCi to load extra object files, archives or DLLs needed by your current session. Restart GHCi, specifying the missing library using the -L/path/to/object/dir and -lmissinglibname flags, or simply by naming the relevant files on the GHCi command line. Alternatively, this link failure might indicate a bug in GHCi. If you suspect the latter, please send a bug report to: glasgow-haskell-bugs(a)haskell.org }}} Also affects GHC 8.0.2 and 7.10. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13997> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0