June 2014 - ghc-tickets

[GHC] #8150: Foreign export requires redundant type signature
by GHC 19 Mar '15

19 Mar '15

#8150: Foreign export requires redundant type signature -------------------------+------------------------------------------------- Reporter: | Owner: evincarofautumn | Status: new Type: bug | Milestone: Priority: | Version: 7.6.3 normal | Operating System: Unknown/Multiple Component: | Type of failure: Incorrect warning at Compiler | compile-time Keywords: | Test Case: Architecture: | Blocking: Unknown/Multiple | Difficulty: | Unknown | Blocked By: | Related Tickets: | -------------------------+------------------------------------------------- With a source file such as: {{{ module FFITest where foreign export ccall "function" function :: IO () function :: IO () function = putStrLn "Hello, world!" }}} The second `function :: IO ()` signature is required to satisfy `-fwarn- missing-signatures`, even though a signature is already present in the `foreign export` declaration, and the types are required to be the same. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8150> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #8315: Improve specialized Hoopl module
by GHC 18 Mar '15

18 Mar '15

#8315: Improve specialized Hoopl module ------------------------------------+------------------------------------- Reporter: jstolarek | Owner: Type: task | Status: new Priority: low | Milestone: Component: Compiler | 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: | ------------------------------------+------------------------------------- GHC uses Hoopl to perform some of the Cmm-to-Cmm optimization passes, but instead of relying directly on Hoopl library GHC uses its own specialized module (see [[GhcFile(compiler/cmm/Hoopl/Dataflow.hs)]]). Interface of that module, as well as its internal implementation can be improved. Right now interface places responsibility of maintaining some of the invariants on user. It also ignores differences between forward and backward analysis. [http://ghc.haskell.org/trac/ghc/wiki/Hoopl/Cleanup This wiki page] gives full detail about proposed design changes. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8315> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

[GHC] #9122: Make Lint check for bad uses of `unsafeCoerce`
by GHC 10 Mar '15

10 Mar '15

#9122: Make Lint check for bad uses of `unsafeCoerce` ------------------------------------+------------------------------------- Reporter: simonpj | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.2 Keywords: | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- I think it would be a great idea for Core Lint to check for uses of `unsafeCoerce` that don't obey the rules. It won't catch all cases, of course, but it would have caught #9035. Specficially, look for: * Coercions between lifted and unboxed types * Coercion between unboxed types of different sizes * Coercion between unboxed ints and floats. Would anyone like to make a patch for this? Anything that can be checked by Core Lint, should be checked! I'm afraid I don't know where to look for the reason for the int/float difficulty. I'd write a tiny function that exhibits the unsafe conversion and look the code it generates. Simon -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9122> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 17

[GHC] #8780: abs for IEEE floating point is slightly wrong.
by GHC 08 Mar '15

08 Mar '15

#8780: abs for IEEE floating point is slightly wrong. ------------------------------------+------------------------------------- Reporter: augustss | Owner: Type: bug | Status: new Priority: low | Milestone: Component: Compiler | Version: 7.6.3 Keywords: | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- Evaluating abs(-0) gives the answer -0. This unexpected since it breaks invariants like 1/(abs x) >= 0. So abs(-0) should be 0. This is also the norm for other languages. Together with this change, signum should also be changed so signum(-0) is -0. This maintains the invariant abs x * signum x == x. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8780> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #8274: Core pretty-printer doesn't print # on unboxed literals
by GHC 03 Mar '15

03 Mar '15

#8274: Core pretty-printer doesn't print # on unboxed literals ------------------------------------+------------------------------------- Reporter: jstolarek | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.7 Keywords: | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: Other Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- I just noticed that Core pretty printer produces something like this (this is part of expected output for T3772 in testsuite): {{{ case GHC.Prim.<=# n# 0 of _ { __DEFAULT -> xs n#; 1 -> GHC.Tuple.() } }}} Both `0` (in scrutinee) and `1` (one of branch alternatives) are unboxed values, so they should be printed as `0#` and `1#` respectively. I haven't checked whether unboxed literals of other types (`Word`, `Double`, `Float`) are printed in that way, but I would guess that yes. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8274> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 9

[GHC] #8342: System.Directory.getPermissions does not work well on Windows 7
by GHC 02 Mar '15

02 Mar '15

#8342: System.Directory.getPermissions does not work well on Windows 7 ---------------------------------------+--------------------------------- Reporter: freizl | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: libraries/directory | Version: 7.4.1 Keywords: | Operating System: Windows Architecture: x86_64 (amd64) | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ---------------------------------------+--------------------------------- It works well for following code {{{ System.Directory.getPermissions "/path/to/somewhere" }}} However getting error with this: {{{ System.Directory.getPermissions "/path/to/somewhere/" *** Exception: getPermissions: does not exist (No such file or directory) }}} The only difference is the path name. It gets error when having extra trailing slash. It looks wired. Thanks your help. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8342> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 11

[GHC] #9044: createDirectoryIfMissing does not fail on unwritable parent dir
by GHC 02 Mar '15

02 Mar '15

#9044: createDirectoryIfMissing does not fail on unwritable parent dir -------------------------------------+------------------------------------- Reporter: duncan | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: | Version: 7.8.2 libraries/directory | Operating System: MacOS X Keywords: | Type of failure: Incorrect result Architecture: Unknown/Multiple | at runtime Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | -------------------------------------+------------------------------------- On OSX, using `createDirectoryIfMissing` to create a new directory inside a directory for which we do not have write permissions does not throw any exception. Of course it should throw an exception, as it does under Linux. {{{ $ mkdir unwritable $ chmod ugo-w unwritable $ ghci > System.Directory.createDirectoryIfMissing True "unwritable/tst" > }}} On OSX (10.9.2) this does not fail. Under Linux (albeit with 7.6.3) this does throw an exception as expected: {{{ *** Exception: unwritable/tst: createDirectory: permission denied (Permission denied) }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9044> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #8896: ghci fails on Arm - "Illegal instruction"
by GHC 01 Mar '15

01 Mar '15

#8896: ghci fails on Arm - "Illegal instruction" ----------------------------+------------------------------- Reporter: Ansible | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.1-rc2 Keywords: arm ghci | Operating System: Linux Architecture: arm | Type of failure: GHCi crash Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ----------------------------+------------------------------- With the help of qemu compiled ghc 7.8.1-rc2 for raspberry pi. Ghc is successful, and ghci compiles, but ghci gets an 'Illegal instruction' error. A sample session: ccrma@satellite:~$ ghc --version The Glorious Glasgow Haskell Compilation System, version 7.8.0.20140228 ccrma@satellite:~$ ghci --version The Glorious Glasgow Haskell Compilation System, version 7.8.0.20140228 ccrma@satellite:~$ ghci GHCi, version 7.8.0.20140228: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... Illegal instruction ccrma@satellite:~$ -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8896> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 10

[GHC] #8189: Default to infinite stack size?
by GHC 24 Feb '15

24 Feb '15

#8189: Default to infinite stack size? ------------------------------------+------------------------------------- Reporter: nh2 | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.6.3 Keywords: | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: Runtime crash Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- http://www.haskell.org/pipermail/haskell-cafe/2013-August/108570.html It looks like any code of the form {{{ ... = do x1 <- m1 x2 <- m2 [do something with x1 and m2] }}} will stack overflow when used with a strict monad (like IO) and sufficiently large data to deal with (a 1m element list is enough to exceed the default 8MB stack size limit). Other examples include: {{{ sequence (m:ms) = do x <- m xs <- sequence ms return (x:xs) }}} and {{{ liftM2, liftM3, ... }}} By slightly modifying your program to do the same thing via heap allocation (e.g. using a difference list in sequence), a stack-overflow rejected program turns into a valid program. This means that valid and idiomatic code like {{{ xs <- replicateM n someIOAction xs <- mapM someIOAction [1..n] }}} will eventually crash, which has an impact on practically all Haskell software and libraries out there. For examples, see https://github.com/ndmitchell/shake/blob/e0e0a43/Development/Shake/Database… https://github.com/haskell-suite/haskell-src-exts/issues/27 as well as the mapM usages in cabal, haddock, any database binding that allows querying a lists of results, etc. While some argue that ''"you shouldn't use lists of that size"'', ''"allocating a 150 MB list is wrong"'' or that they ''"carefully avoid any use of sequence/mapM due to the problem"'' pointed out above, I think that collecting a million results of a strict monadic action into a linked list using standard functions like mapM is a common pattern and valid use case. The stack based evaluation that GHC performs is an optimization and should be transparent to the user; it should not re-define some data to live in a fixed-size memory region, and it should work just as well as it does in GHCi. It seems that a default infinite stack size would mitigate this. A drawback might be that the stack space overflows we get so far to quickly spot a "wrong" program would not indicate certain errors any more. I currently doubt the usefulness of this debugging method, since it does not work reliably in the presence of optimization, doing the same thing via heap allocation makes the programs "correct", and it rejects valid programs if the input data is large enough. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8189> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 35

[GHC] #9103: hackage's type-level-0.2.4 fails to compile
by GHC 23 Feb '15

23 Feb '15

#9103: hackage's type-level-0.2.4 fails to compile -------------------------------------+------------------------------------- Reporter: slyfox | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler (Type | Version: 7.8.2 checker) | Operating System: Unknown/Multiple Keywords: | Type of failure: GHC rejects Architecture: Unknown/Multiple | valid program Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | -------------------------------------+------------------------------------- Attached trimmed-down dependencyless archive. Steps to reproduce: {{{ $ tar -xzf type-level-0.2.4.tar.gz $ ghci -hide-all-packages -package=base -package=template-haskell Data/TypeLevel.hs }}} The error on 7.8.2 and in -HEAD is: {{{ [6 of 8] Compiling Data.TypeLevel.Num.Ops ( Data/TypeLevel/Num/Ops.hs, interpreted ) Data/TypeLevel/Num/Ops.hs:90:10: Illegal instance declaration for ‘Succ' (x, x) (x, x) D0 D0 True’ The liberal coverage condition fails in class ‘Succ'’ for functional dependency: ‘yh yl yz -> xh xl’ Reason: lhs types ‘D0’, ‘D0’, ‘True’ do not jointly determine rhs types ‘(x, x)’, ‘(x, x)’ In the instance declaration for ‘Succ' (x, x) (x, x) D0 D0 True’ ... }}} Typechecked fine on ghc-7.6.3 {{{ [1 of 8] Compiling Data.TypeLevel.Num.Reps ( Data/TypeLevel/Num/Reps.hs, interpreted ) [2 of 8] Compiling Data.TypeLevel.Num.Sets ( Data/TypeLevel/Num/Sets.hs, interpreted ) [3 of 8] Compiling Data.TypeLevel.Num.Aliases.TH ( Data/TypeLevel/Num/Aliases/TH.hs, interpreted ) [4 of 8] Compiling Data.TypeLevel.Num.Aliases ( Data/TypeLevel/Num/Aliases.hs, interpreted ) Generating and compiling a zillion numerical type aliases, this might take a while [5 of 8] Compiling Data.TypeLevel.Bool ( Data/TypeLevel/Bool.hs, interpreted ) [6 of 8] Compiling Data.TypeLevel.Num.Ops ( Data/TypeLevel/Num/Ops.hs, interpreted ) [7 of 8] Compiling Data.TypeLevel.Num ( Data/TypeLevel/Num.hs, interpreted ) [8 of 8] Compiling Data.TypeLevel ( Data/TypeLevel.hs, interpreted ) Ok, modules loaded: Data.TypeLevel, Data.TypeLevel.Num, Data.TypeLevel.Bool, Data.TypeLevel.Num.Reps, Data.TypeLevel.Num.Aliases, Data.TypeLevel.Num.Sets, Data.TypeLevel.Num.Ops, Data.TypeLevel.Num.Aliases.TH. *Data.TypeLevel> }}} I have failed to decipher an error message, but looks like it tries to instantiate a thing it should not. My apologies if it's not-a-bug. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9103> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 18