May 2014 - ghc-tickets

[GHC] #8146: Library package cannot be found
by GHC 20 Jun '14

20 Jun '14

#8146: Library package cannot be found --------------------------------+------------------------------------------ Reporter: conradwt | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.6.3 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Installing GHC failed Unknown/Multiple | Test Case: Difficulty: Unknown | Blocking: Blocked By: | Related Tickets: | --------------------------------+------------------------------------------ {{{ ./sync-all get warning: libffi-tarballs already present; omitting == libffi-tarballs: running git config --local core.ignorecase true warning: utils/hsc2hs already present; omitting == utils/hsc2hs: running git config --local core.ignorecase true warning: utils/haddock already present; omitting == utils/haddock: running git config --local core.ignorecase true == running git clone https://github.com/ghc/packages/array.git libraries/array Cloning into 'libraries/array'... fatal: repository 'https://github.com/ghc/packages/array.git/' not found git failed: 32768 at ./sync-all line 119. == Checking for old haddock repo == Checking for old binary repo == Checking for old mtl repo == Checking for old Cabal repo == Checking for old time from tar ball }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8146> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #8667: sync-all doesn't work properly if you run from a fork on github
by GHC 20 Jun '14

20 Jun '14

#8667: sync-all doesn't work properly if you run from a fork on github ------------------------------------+------------------------------------- Reporter: schyler | Owner: hvr Type: bug | Status: new Priority: normal | Milestone: Component: Trac & Git | 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: | ------------------------------------+------------------------------------- The intuitive behaviour when working with projects on github is to make your own fork so that you can use the online interface (or GitHub for {windows,osx} apps). One would assume that this behaviour was endorsed because of the ~130 forks to the ghc repo, however forking the repo actually causes ./sync-all to stop working (it asks for a username and password to whomever@ github.com) It seems the only way for a person to work with the project from github is to clone directly from the original repo. This sounds like a bug. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8667> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #9138: Allow users to edit their own tickets
by GHC 18 Jun '14

18 Jun '14

#9138: Allow users to edit their own tickets ------------------------------------+------------------------------------- Reporter: nomeata | Owner: hvr Type: bug | Status: new Priority: normal | Milestone: Component: Trac & Git | 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: | ------------------------------------+------------------------------------- Non-power-users can edit their own comments, but not their tickets. That is annoying when one makes mistakes in the syntax. There is a history of changes, so it really wouldn't be a problem to allow that. The required permissions according to [TracPermissions] seems to be `TICKET_EDIT_DESCRIPTION`. `TICKET_EDIT_CC` would also be useful (#8881), as I can put people on CC when I create a ticket, but I cannot change that later. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9138> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #9097: Change GHC.Exts.Any to a type family
by GHC 11 Jun '14

11 Jun '14

#9097: Change GHC.Exts.Any to a type family ------------------------------------+------------------------------------- Reporter: goldfire | Owner: Type: task | 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 just had this slightly alarming interchange with GHCi: {{{ Prelude> import Data.Type.Equality Prelude Data.Type.Equality> import GHC.Exts Prelude Data.Type.Equality GHC.Exts> :kind! ((Any :: Bool) == (Any :: Bool)) ((Any :: Bool) == (Any :: Bool)) :: Bool = 'False }}} After staring at the result in disbelief, I figured out why. The instance for `==` at kind `Bool` looks like this: {{{ type family EqBool a b where EqBool False False = True EqBool True True = True EqBool a b = False type instance (a :: Bool) == (b :: Bool) = EqBool a b }}} Well, `Any` isn't `False`, `Any` isn't `True`, so `Any == Any` must be `False`! The solution to this, of course, is to make `Any` a type family, not a datatype. Then, it wouldn't be apart from the equations in `EqBool`. I believe this idea has been floated previously but was not implemented as it would have disturbed !TypeLits and/or singletons. These libraries have been updated, and it's time. I'm happy to do this myself in due course. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9097> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 8

[GHC] #9062: Role: ~# should be ~R#
by GHC 11 Jun '14

11 Jun '14

#9062: Role: ~# should be ~R# ------------------------------------+--------------------------------- Reporter: conal | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.2 Keywords: | Operating System: MacOS X Architecture: Unknown/Multiple | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+--------------------------------- The following coercion showed up in Core. Richard Eisenberg pointed out that the `~#` should be `~R#`. {{{ <Tree N0 Int>_R -> NTCo:Sum[0] <Int>_R :: (Tree N0 Int -> Sum Int) ~# (Tree N0 Int -> Int) }}} I can dig up more information if needed. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9062> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #9127: Don't warn about pattern-bindings of the form `let !_ = rhs`
by GHC 09 Jun '14

09 Jun '14

#9127: Don't warn about pattern-bindings of the form `let !_ = rhs` ------------------------------------+------------------------------------- Reporter: refold | 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: | ------------------------------------+------------------------------------- With GHC 7.8.2, code like {{{ let !_ = rqPostParams rq let !_ = rqParams rq }}} triggers the following warnings: {{{ src/Snap/Internal/Test/RequestBuilder.hs:150:13: Warning: This pattern-binding binds no variables: !_ = rqPostParams rq src/Snap/Internal/Test/RequestBuilder.hs:151:13: Warning: This pattern-binding binds no variables: !_ = rqParams rq }}} I think that `let !_ = rhs` shouldn't trigger a warning, just like `let _ = rhs` doesn't. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9127> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 15

[GHC] #8753: Import constructor but not the data type
by GHC 09 Jun '14

09 Jun '14

#8753: Import constructor but not the data type ------------------------------------+------------------------------------- Reporter: andreas.abel | Owner: Type: feature request | Status: new Priority: normal | 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: | ------------------------------------+------------------------------------- How do I import a constructor but not its data type? The constructor might have the name of a data type which I do not want to bring into scope. Use case: Library (A) defines a parametrized data type (D a) which I want to use in my code (B) in instantiated form (D Int) but with the same name (D). I want to pattern match against inhabitants of type D, so I need the constructor(s) in scope. Here is how it could look like if import lists let me specify whether I want to import a type or a constructor with name "D". module A where data D a = D a module B where import A (constructor D) import qualified A D = A.D Int f :: D -> Int f (D x) = x Haskell has different name spaces for type and constructors, but does not let the user talk about these name spaces where it matters, namely in import/export lists. At least I found no documentation instructing me how to do this. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8753> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 11

[GHC] #9050: Panic when compiling cmm file together with -outputdir
by GHC 09 Jun '14

09 Jun '14

#9050: Panic when compiling cmm file together with -outputdir -----------------------------------+--------------------------------------- Reporter: Yuras | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.2 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Compile-time crash Unknown/Multiple | Test Case: Difficulty: Unknown | Blocking: Blocked By: | Related Tickets: | -----------------------------------+--------------------------------------- To reproduce: {{{ $ touch test.cmm $ ghc -c test.cmm -outputdir=build ghc-stage2: panic! (the 'impossible' happened) (GHC version 7.9.20140428 for x86_64-unknown-linux): hscCmmFile: no_mod }}} The panic is declared in `main/HscMain.hs:hscCompileCmmFile` in place of current module name, but forced in `main/CodeOutput.lhs:outputForeignStubs` by `doesFileExist`. It works without `-outputdir` because in that case module name is not used to create path to stub file, see `main/Finder.lhs:mkStubPaths`. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9050> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #9080: GHC error on Win 8.1 64bit
by GHC 09 Jun '14

09 Jun '14

#9080: GHC error on Win 8.1 64bit ----------------------------------+--------------------------------------- Reporter: maun | Owner: Type: bug | Status: new Priority: normal | Milestone: 7.8.3 Component: Compiler | Version: 7.8.2 Keywords: | Operating System: Windows Architecture: x86_64 (amd64) | Type of failure: Compile-time crash Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ----------------------------------+--------------------------------------- Hi, i get the following error when building GHCJS: {{{ Loading package conduit-1.1.1 ... linking ... done. Loading package yaml-0.8.8.2 ... ghc.exe: panic! (the 'impossible' happened) (GHC version 7.8.2 for x86_64-unknown-mingw32): loadObj "C:\\Users\\Manuel\\AppData\\Roaming\\cabal\\x86_64 -windows-ghc- 7.8.2\\yaml-0.8.8.2\\HSyaml-0.8.8.2.o": failed Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug ghc.exe: Unknown PEi386 section name `.text.unlikely' (while processing: C:\User s\Manuel\AppData\Roaming\cabal\x86_64-windows- ghc-7.8.2\yaml-0.8.8.2\HSyaml-0.8. 8.2.o) cabal: Error: some packages failed to install: ghcjs-0.1.0 failed during the building phase. The exception was: ExitFailure 1 }}} Also reported this for ghcjs (https://github.com/ghcjs/ghcjs/issues/170) I am not using the haskell plattform, but the 7.8.2 binaries + MinGW and MSYS. I also tried mingw-w64 but this made no difference. I am also using the recent ghcjs cabal branch. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9080> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #9155: allocateRegsAndSpill: Cannot read from uninitialized register %vI_s154O (GHC version 7.8.2 for x86_64-unknown-linux)
by GHC 09 Jun '14

09 Jun '14

#9155: allocateRegsAndSpill: Cannot read from uninitialized register %vI_s154O (GHC version 7.8.2 for x86_64-unknown-linux) ------------------------------------+------------------------------------- Reporter: slyfox | 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: | ------------------------------------+------------------------------------- When building all world with '''-O2''' I've stuck on wxhaskell (all versions): {{{ [11 of 16] Compiling Graphics.UI.WX.Controls ( src/Graphics/UI/WX/Controls.hs, dist/build/Graphics/UI/WX/Controls.o ) <no location info>: ghc: panic! (the 'impossible' happened) (GHC version 7.8.2 for x86_64-unknown-linux): allocateRegsAndSpill: Cannot read from uninitialized register %vI_s154O }}} I've stripped it down to minimal complete example (but it significantly changed error message) {{{ ghc: panic! (the 'impossible' happened) (GHC version 7.8.2 for x86_64-unknown-linux): RegAlloc.Liveness.computeLivenss SCCs aren't in reverse dependent order bad blockId c1tU [NONREC c1uc: movq $-8,%rbx andq $main:M.AlignLeft{v rb}_closure+1,%rbx addq $8,%rbp jmp *(%rbx), NONREC ... }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9155> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 10