April 2015 - ghc-tickets

[GHC] #9607: Type checking regression between GHC 7.6 and 7.8
by GHC 26 May '18

26 May '18

#9607: Type checking regression between GHC 7.6 and 7.8 -------------------------------------+------------------------------------- Reporter: jstolarek | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler (Type | Version: 7.8.3 checker) | Operating System: Keywords: | Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: GHC Difficulty: Unknown | rejects valid program Blocked By: | Test Case: Related Tickets: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- Jason McCarty [http://www.haskell.org/pipermail/haskell- cafe/2014-September/116076.html reported on Haskell-cafe] that this code used to work with GHC 7.6 but in GHC 7.8 it requires `AllowAmbiguousTypes`: {{{#!hs -- The code below is simplified from code that computes a tensor product of -- a tensor with an identity matrix whose size is determined from the -- shapes of the input and output tensors. {-# LANGUAGE DataKinds, KindSignatures, TypeFamilies, TypeOperators #-} --{-# LANGUAGE AllowAmbiguousTypes #-} module Tensors where import GHC.TypeLits type family (as :: [Nat]) ++ (bs :: [Nat]) :: [Nat] type instance '[] ++ bs = bs type instance (a ': as) ++ bs = a ': (as ++ bs) data Tensor (s :: [Nat]) = Tensor -- content elided -- apparently GHC reduces (++) enough to see that n is determined leftUnit :: Tensor s -> Tensor ('[n, n] ++ s) leftUnit Tensor = Tensor -- accepted in 7.6, not accepted in 7.8 without AllowAmbiguousTypes rightUnit :: Tensor s -> Tensor (s ++ '[n, n]) rightUnit Tensor = Tensor -- also accepted without AllowAmbiguousTypes outsideUnit :: Tensor s -> Tensor ('[n] ++ s ++ '[n]) outsideUnit Tensor = Tensor useleftUnit :: Tensor '[1,1,2] useleftUnit = leftUnit Tensor -- type of Tensor is inferred userightUnit :: Tensor '[1,2,2] userightUnit = rightUnit (Tensor :: Tensor '[1]) -- type must be provided }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9607> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 10

[GHC] #9980: TcS monad is too heavy
by GHC 24 May '18

24 May '18

#9980: TcS monad is too heavy -------------------------------------+------------------------------------- Reporter: simonpj | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.4 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Blocked By: Test Case: | Related Tickets: Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- The typehcecker's constraint-solving monad, `TcS`, is simply built on top of `TcM`, but it doesn't use most of `TcM`'s facilities. That was fine while `TcS` was only called from the typechecker, but now (as part of [wiki:PatternMatchCheck fixing pattern-match overlap checking]) we invoke it from the desugarer. It seems quite wrong to construct a vast, and unnecessary `TcM` context just to invoke `TcS`. Better: make `TcS` into its own monad, with its own `TcSLclEnv`, built on `IOEnv` like the others. Main objection: the plugins mechanism exposes `unsafeTcPluginTcM`, which would become unavailable. But it it used? -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9980> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 8

[GHC] #10089: feature: warn about unused data definitions (with typeclass instances)
by GHC 23 May '18

23 May '18

#10089: feature: warn about unused data definitions (with typeclass instances) -------------------------------------+------------------------------------- Reporter: slyfox | Owner: Type: feature | Status: new request | Milestone: Priority: normal | Version: 7.8.4 Component: Compiler | Operating System: Unknown/Multiple Keywords: | Type of failure: None/Unknown Architecture: | Blocked By: Unknown/Multiple | Related Tickets: Test Case: | Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- Let's consider an example: {{{#!hs -- M.hs: module M () where -- ghc knows about uselessness of those data D = D -- but doesn't about this: data E = E deriving (Show, Read) }}} '''ghc''' warns about useless '''D''', but not '''E''': {{{ $ ghc-7.8.4 -Wall -c M.hs -fforce-recomp M.hs:5:1: Warning: Defined but not used: type constructor or class ‘D’ M.hs:5:10: Warning: Defined but not used: data constructor ‘D’ }}} There is no way to refer to '''E''' type outside or call it's unstances, correct? I would expect no code should be generated for this module. '''ghc''' could also warn about defined, but not used instances for types, that are not exported. Some background on where those stray types come from: 1. programmer usually adds a set of data types into their program (say, 10-15 types to represent an AST) 2. then annotates them with 'deriving (Read, Show)' convenience instances 3. (months later) amends AST representation a bit and leaves leftover data types not used anywhere Does that make sense? Thanks! -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10089> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 8

[GHC] #9918: GHC chooses an instance between two overlapping, but cannot resolve a clause within the similar closed type family
by GHC 23 May '18

23 May '18

#9918: GHC chooses an instance between two overlapping, but cannot resolve a clause within the similar closed type family -------------------------------------+------------------------------------- Reporter: qnikst | Owner: Type: bug | Status: new Priority: normal | Milestone: 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: -------------------------------------+------------------------------------- We hoped that closed type families could replace overlapping instances. It is not currently the case: GHC can resolve overlapping instances but cannot resolve the similar closed type family. I'm attaching a minimized example of program where it can be observed. This is reproducible on ghc-7.8.3 and HEAD. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9918> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 21

[GHC] #10241: BlockedIndefinitelyOnMVar thrown to the thread which is not blocked indefinitely
by GHC 21 May '18

21 May '18

#10241: BlockedIndefinitelyOnMVar thrown to the thread which is not blocked indefinitely -------------------------------------+------------------------------------- Reporter: asukamirai | Owner: simonmar Type: bug | Status: new Priority: normal | Milestone: Component: Runtime | Version: 7.8.3 System | Operating System: Unknown/Multiple Keywords: | Type of failure: Incorrect result Architecture: x86_64 | at runtime (amd64) | Blocked By: Test Case: | Related Tickets: Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- BlockedIndefinatelyOnMVar exception is thrown to the main thread in below source code although the thread is not blocked indefinitely. {{{#!hs module Main where import qualified Control.Concurrent.MVar as MV import qualified Control.Concurrent as CC import qualified Control.Exception as E main :: IO () main = do -- call this thread "threadA" mvar1 <- MV.newEmptyMVar :: IO (MV.MVar ()) mvar2 <- MV.newEmptyMVar :: IO (MV.MVar ()) _ <- CC.forkIO $ do -- call this thread "threadB" MV.takeMVar mvar1 `E.catch` errorHandler1 putStrLn "after error catch" CC.threadDelay 1000000 MV.putMVar mvar2 () putStrLn "after putMVar" MV.readMVar mvar2 `E.catch` errorHandler2 putStrLn "after readMVar" CC.threadDelay 5000000 where errorHandler1 :: E.BlockedIndefinitelyOnMVar -> IO () errorHandler1 e = putStrLn $ "errorHandler1 : " ++ show e errorHandler2 :: E.BlockedIndefinitelyOnMVar -> IO () errorHandler2 e = putStrLn $ "errorHandler2 : " ++ show e }}} Save above as "mvar.hs" and run by ghc as below. {{{ > runhaskell mvar.hs errorHandler1 : thread blocked indefinitely in an MVar operation errorHandler2 : thread blocked indefinitely in an MVar operation after error catch after readMVar after putMVar }}} BlockedIndefinitelyOnMVar thrown for mvar1 is correct. It will be caught by errorHandler1 and "threadB" can continue to put the value to mvar2. It means that "threadA" can wait for the value of mvar2 and it is not blocked indefinately. However, BlockedIndefinitelyOnMVar is thrown for mvar2 on "threadA" before "threadB" puts value to the mvar2. I think it is incorrect. ---- I tested another case that adding "CC.threadDelay 10000000" before "readMVar" as below. {{{#!hs module Main where import qualified Control.Concurrent.MVar as MV import qualified Control.Concurrent as CC import qualified Control.Exception as E main :: IO () main = do -- call this thread "threadA" mvar1 <- MV.newEmptyMVar :: IO (MV.MVar ()) mvar2 <- MV.newEmptyMVar :: IO (MV.MVar ()) _ <- CC.forkIO $ do -- call this thread "threadB" MV.takeMVar mvar1 `E.catch` errorHandler1 putStrLn "after error catch" CC.threadDelay 1000000 MV.putMVar mvar2 () putStrLn "after putMVar" CC.threadDelay 10000000 -- <-- this line is added MV.readMVar mvar2 `E.catch` errorHandler2 putStrLn "after readMVar" CC.threadDelay 5000000 where errorHandler1 :: E.BlockedIndefinitelyOnMVar -> IO () errorHandler1 e = putStrLn $ "errorHandler1 : " ++ show e errorHandler2 :: E.BlockedIndefinitelyOnMVar -> IO () errorHandler2 e = putStrLn $ "errorHandler2 : " ++ show e }}} And it will run correctly (BlockedIndefinitelyOnMVar is not thrown for mvar2). {{{ > runhaskell mvar.hs errorHandler1 : thread blocked indefinitely in an MVar operation after error catch after putMVar after readMVar }}} ---- I found this behavior is same on STM / BlockedIndefinitelyOnSTM. {{{#!hs module Main where import qualified Control.Concurrent.STM as STM import qualified Control.Concurrent as CC import qualified Control.Exception as E main :: IO () main = do tmv1 <- STM.newEmptyTMVarIO :: IO (STM.TMVar ()) tmv2 <- STM.newEmptyTMVarIO :: IO (STM.TMVar ()) _ <- CC.forkIO $ do STM.atomically (STM.takeTMVar tmv1) `E.catch` errorHandler1 putStrLn "after error catch" CC.threadDelay 1000000 STM.atomically $ STM.putTMVar tmv2 () putStrLn "after putTMVar" STM.atomically (STM.readTMVar tmv2) `E.catch` errorHandler2 putStrLn "after readTMVar" CC.threadDelay 5000000 where errorHandler1 :: E.BlockedIndefinitelyOnSTM -> IO () errorHandler1 e = putStrLn $ "errorHandler1 : " ++ show e errorHandler2 :: E.BlockedIndefinitelyOnSTM -> IO () errorHandler2 e = putStrLn $ "errorHandler2 : " ++ show e }}} {{{ > runhaskell stm.hs errorHandler1 : thread blocked indefinitely in an STM transaction errorHandler2 : thread blocked indefinitely in an STM transaction after error catch after readTMVar after putTMVar }}} ---- I tested this in below versions/OSs and got same result (exception thrown for mvar2/tmv2). ghc7.8.3 on Windows7 ghc7.8.3 on lubuntu14.04 on VirtualBox on Windows7 ghc7.8.4 on lubuntu14.04 on VirtualBox on Windows7 ghc7.10.1 on lubuntu14.04 on VirtualBox on Windows7 Similar report https://ghc.haskell.org/trac/ghc/ticket/8804 found but not the same. (In this case, the reference to the MVar is not weak) -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10241> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #9640: Invalid object in processHeapClosureForDead(): 2130762223
by GHC 19 May '18

19 May '18

#9640: Invalid object in processHeapClosureForDead(): 2130762223 -------------------------------------+------------------------------------- Reporter: Fuuzetsu | Owner: Type: bug | Status: new Priority: normal | Milestone: 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: -------------------------------------+------------------------------------- I just tried using {{{-hb}}} profiling for Yi and hit {{{ yi: internal error: Invalid object in processHeapClosureForDead(): 2130762223 (GHC version 7.8.3 for x86_64_unknown_linux) Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug Aborted }}} Unfortunately I don't have an easy way to replicate. In my case it's to check out HEAD version of Yi, build with profiling and run {{{ [nix-shell:~/programming/yi/yi]$ TERM=xterm yi_datadir=. dist/build/yi/yi -fpango --as=emacs -y /tmp /tmp/long +RTS -p -hb }}} where {{{/tmp/long}}} here is just a file with 1MB of the letter {{{A}}} repeated on a single line. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9640> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #9840: Permit empty closed type families
by GHC 18 May '18

18 May '18

#9840: Permit empty closed type families -------------------------------------+------------------------------------- Reporter: adamgundry | Owner: Type: feature request | Status: new Priority: normal | Milestone: Component: Compiler (Type | Version: 7.8.3 checker) | Operating System: Keywords: | Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: Difficulty: Unknown | None/Unknown Blocked By: | Test Case: Related Tickets: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- At the moment, closed type families without any equations fail with a parse error. In addition, they cannot be created by TH (see #8028). Would it be possible to permit these instead? My use case is in my typechecker plugin for units of measure, where I want to add new type-level operators without any equational theory (because it will be supplied by the plugin) and without users having the ability to introduce their own type family instances. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9840> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 13

[GHC] #10078: tcPluginStop of a type checker plugin is not called if an error occurs
by GHC 18 May '18

18 May '18

#10078: tcPluginStop of a type checker plugin is not called if an error occurs -------------------------------------+------------------------------------- Reporter: jbracker | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.11 (Type checker) | Operating System: Linux Keywords: | Type of failure: Incorrect result Architecture: x86_64 | at runtime (amd64) | Blocked By: Test Case: | Related Tickets: Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- When a module using a type checker plugin produces a compiler error the clean up function `tcPluginStop` of the plugin is not called. I am not sure if this is intended, but according to the description of the wiki page (Plugins/TypeChecker) this should always be called. === Test plugin `MyPlugin.hs`: {{{#!hs module MyPlugin ( plugin ) where import Plugins import TcRnTypes import TcPluginM plugin :: Plugin plugin = defaultPlugin { tcPlugin = \clos -> Just $ TcPlugin { tcPluginInit = tcPluginIO $ putStrLn ">>> Plugin Init" , tcPluginSolve = \_ _ _ _ -> do tcPluginIO $ putStrLn ">>> Plugin Solve" return $ TcPluginOk [] [] , tcPluginStop = \_ -> tcPluginIO $ putStrLn ">>> Plugin Stop" } } }}} === Minimal example (with type error) `Main.hs`: {{{#!hs {-# OPTIONS_GHC -fplugin MyPlugin #-} module Main where main :: (Monad m) => m () main = do return 1 }}} Compiling this will lead to the following output: {{{ $ ~/ghc/inplace/bin/ghc-stage2 --make -package ghc-7.11.20150209 -dynamic Main.hs [2 of 2] Compiling Main ( Main.hs, Main.o ) >>> Plugin Init >>> Plugin Solve >>> Plugin Solve >>> Plugin Solve Main.hs:6:10: Could not deduce (Num ()) arising from the literal ‘1’ from the context: Monad m bound by the type signature for: main :: Monad m => m () at Main.hs:4:9-25 In the first argument of ‘return’, namely ‘1’ In a stmt of a 'do' block: return 1 In the expression: do { return 1 } }}} Which means `tcPluginStop` was _not_ called. === Minimal example (without type error) `Main.hs`: {{{#!hs {-# OPTIONS_GHC -fplugin MyPlugin #-} module Main where main :: (Monad m) => m () main = do return () }}} Compiling this will lead to the following output: {{{ $ ~/ghc/inplace/bin/ghc-stage2 --make -package ghc-7.11.20150209 -dynamic Main.hs [2 of 2] Compiling Main ( Main.hs, Main.o ) [MyPlugin changed] >>> Plugin Init >>> Plugin Solve >>> Plugin Solve >>> Plugin Stop Linking Main ... }}} Which means `tcPluginStop` _was_ called. === Possible solution As far as I can see, the solution to this should be to change the plugin code at the bottom of `typechecker/TcRnDriver.hs` from {{{#!hs withTcPlugins :: HscEnv -> TcM a -> TcM a withTcPlugins hsc_env m = do plugins <- liftIO (loadTcPlugins hsc_env) case plugins of [] -> m -- Common fast case _ -> do (solvers,stops) <- unzip `fmap` mapM startPlugin plugins res <- updGblEnv (\e -> e { tcg_tc_plugins = solvers }) m mapM_ runTcPluginM stops return res where startPlugin (TcPlugin start solve stop) = do s <- runTcPluginM start return (solve s, stop s) }}} to {{{#!hs withTcPlugins :: HscEnv -> TcM a -> TcM a withTcPlugins hsc_env m = do plugins <- liftIO (loadTcPlugins hsc_env) case plugins of [] -> m -- Common fast case _ -> do (solvers,stops) <- unzip `fmap` mapM startPlugin plugins eitherRes <- tryM $ do updGblEnv (\e -> e { tcg_tc_plugins = solvers }) m mapM_ runTcPluginM stops case eitherRes of Left e -> failM Right res -> return res where startPlugin (TcPlugin start solve stop) = do s <- runTcPluginM start return (solve s, stop s) }}} . I have tried this. It compiles and my minimal example delivers the correct result. Are there any arguments against this change? If not, I would try to commit a patch for this problem sometime this weekend. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10078> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #10077: Providing type checker plugin on command line results in false cyclic import error
by GHC 18 May '18

18 May '18

#10077: Providing type checker plugin on command line results in false cyclic import error -------------------------------------+------------------------------------- Reporter: jbracker | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.11 (Type checker) | Operating System: Linux Keywords: | Type of failure: Incorrect typechecker plugin cycle imports | warning at compile-time Architecture: x86_64 | Blocked By: (amd64) | Related Tickets: Test Case: | Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- I am playing around with the new type checker plugins using the current development branch of GHC. When I supply the plugin module to GHC through the command line I get a cyclic import error message, which I think is false. If I supply the same plugin module using a pragma I do not get an error message. === Minimal Example (Command Line) `MyPlugin.hs`: {{{#!hs module MyPlugin ( plugin ) where import Plugins ( Plugin, defaultPlugin ) plugin :: Plugin plugin = defaultPlugin }}} `Test.hs`: {{{#!hs module Test where main :: IO () main = return () }}} Command line call of GHC: {{{ ~/ghc/inplace/bin/ghc-stage2 \ --make -package ghc-7.11.20150209 \ -fplugin=MyPlugin \ Test.hs }}} Results in the following error {{{ Module imports form a cycle: module ‘MyPlugin’ (./MyPlugin.hs) imports itself }}} which does not seem reasonable to me understand. === Minimal example (pragma) On the other hand, if I change `Test.hs` to the following {{{#!hs {-# OPTIONS_GHC -fplugin MyPlugin #-} module Test where main :: IO () main = return () }}} and calling GHC like this {{{ ~/ghc/inplace/bin/ghc-stage2 \ --make \ -package ghc-7.11.20150209 \ -dynamic \ Test.hs }}} it works. I did not change `MyPlugin.hs`. === Note 1. Using `-fplugin=MyPlugin` vs. `-fplugin MyPlugin` does not make a difference. 1. The command line example behaves the same independent of whether I supply the `-dynamic` flag or not. 1. I had to add the `-dynamic` flag, because otherwise GHC will complain that: {{{ <no location info>: cannot find normal object file ‘./MyPlugin.dyn_o’ while linking an interpreted expression }}} This might be a long shot, but maybe using the `-fplugin` option should imply the `-dynamic` flag? -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10077> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 11

[GHC] #8089: Implementation of GHC.Event.Poll.poll is broken due to bad coercion
by GHC 16 May '18

16 May '18

#8089: Implementation of GHC.Event.Poll.poll is broken due to bad coercion -------------------------------------------+------------------------------- Reporter: merijn | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: libraries/base | Version: 7.6.3 Keywords: patch | Operating System: Architecture: Unknown/Multiple | Unknown/Multiple Difficulty: Easy (less than 1 hour) | Type of failure: Runtime Blocked By: | crash Related Tickets: | Test Case: | Blocking: -------------------------------------------+------------------------------- Due to an unsafe coercion in GHC.Event.Poll.poll the implementation has a bug that on some platforms (.e.g. linux) results in subtly wrong semantics and on other platforms (e.g. OS X) downright crashes the runtime. The offending line is line 95 in GHC/Event/Poll.hsc: {{{ c_poll ptr (fromIntegral len) (fromIntegral (fromTimeout tout)) }}} `fromTimeout` returns a timeout in seconds as a value of type Int. `c_poll` takes a timeout in seconds as a value of type CInt. If the max value of Int is larger than that of CInt, this coercion results in an overflow, producing a negative value. The faulty semantics are in turn caused by passing this negative value as timeout to `c_poll`. This overflow happens on many 64bit platforms where GHC's Int is 64bit, but CInt is often 32bit. The result of passing a negative value to poll() depends on the platform, but is (on all platforms I checked) different from the desired semantics! On linux any negative timeouts are treated as infinite, this means that a poll() with a timeout that should be finite results in an infinitely blocking poll on linux! On OSX any negative timeout other than -1 (which is treated as infinity) is considered an invalid parameter, returning an error and crashing the runtime. A minimal example reproducing the error (on systems where `fromIntegral (maxBound :: Int) > (fromIntegral (maxBound :: CInt) :: Integer)`): {{{ import Control.Concurrent main = threadDelay maxBound }}} Using 64bit GHC on OSX this produces an immediate runtime crash. The attached patch solves the issue by comparing the Int timeout against the max value of CInt and, if necessary, looping multiple `c_poll` calls (with a timeout equal to `maxBound :: CInt`) if they return `0` (which indicates the timeout has expired with no events occurring). See also: https://github.com/merijn/packages- base/commit/6d8ea02a3f6a2cdb82a9ad786a8c4db780b92091 -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8089> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 22