April 2015 - ghc-tickets

[GHC] #9267: Lack of type information in GHC error messages when the liberage coverage condition is unsatisfied
by GHC 25 Jul '15

25 Jul '15

#9267: Lack of type information in GHC error messages when the liberage coverage condition is unsatisfied -------------------------+------------------------------------------------- Reporter: | Owner: danilo2 | Status: new Type: bug | Milestone: Priority: | Version: 7.8.2 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: | -------------------------+------------------------------------------------- Hello! The problem is not a bug related to proper code execution, but it affects the development in such way, we can treat it as a bug. Lets think about such simple code (it can be further simplified of course): {{{#!haskell {-# LANGUAGE FunctionalDependencies #-} {-# LANGUAGE FlexibleInstances #-} {-# LANGUAGE ScopedTypeVariables #-} {-# LANGUAGE UndecidableInstances #-} data UnsafeBase base err val = Value val | Error err | Other (base val) deriving (Show, Eq) class MagicMerge m1 m2 | m1 -> m2 where magicMerge :: m1 a -> m2 a instance MagicMerge (UnsafeBase (UnsafeBase base err) err) (UnsafeBase base err) where magicMerge ma = case ma of Value a -> Value a Error e -> Error e Other o -> o instance MagicMerge (UnsafeBase (UnsafeBase base err1) err2) (UnsafeBase dstBase err1) where magicMerge (ma :: UnsafeBase (UnsafeBase base err1) err2 a) = case ma of Value a -> Value a Error e -> Other $ magicMerge (Error e :: UnsafeBase base err2 a) Other o -> case o of Value a -> Other $ magicMerge (Value a :: UnsafeBase base err2 a) Error e -> Error e Other o' -> Other $ magicMerge (Other o' :: UnsafeBase base err2 a) main = print "help me world!" }}} When trying to compile it using GHC-7.8, we get following error message: {{{#!haskell Illegal instance declaration for ‘MagicMerge (UnsafeBase (UnsafeBase base err1) err2) (UnsafeBase dstBase err1)’ The liberal coverage condition fails in class ‘MagicMerge’ for functional dependency: ‘m1 -> m2’ Reason: lhs type ‘UnsafeBase (UnsafeBase base err1) err2’ does not determine rhs type ‘UnsafeBase dstBase err1’ In the instance declaration for ‘MagicMerge (UnsafeBase (UnsafeBase base err1) err2) (UnsafeBase dstBase err1)’ }}} Which is of course right, but it is completely unhelpfull! When we run the same program against GHC-7.6, we get another message: {{{#!haskell No instance for (MagicMerge (UnsafeBase base err2) dstBase) arising from a use of `magicMerge' Possible fix: add an instance declaration for (MagicMerge (UnsafeBase base err2) dstBase) In the second argument of `($)', namely `magicMerge (Error e :: UnsafeBase base err2 a)' In the expression: Other $ magicMerge (Error e :: UnsafeBase base err2 a) In a case alternative: Error e -> Other $ magicMerge (Error e :: UnsafeBase base err2 a) }}} And it contains a very helpfull information about inferred types. The information is very helpfull, especialy with more complex instances. What is VERY IMPORTANT, when we use the inferred type, the code starts working in both, GHC-7.8 and GHC-7.6. Right now I have to use both versions of GHC to be able to develop my instances faster. So If we add the inferred by GHC-7.6 premise: {{{#!haskell {-# LANGUAGE FunctionalDependencies #-} {-# LANGUAGE FlexibleInstances #-} {-# LANGUAGE ScopedTypeVariables #-} {-# LANGUAGE UndecidableInstances #-} data UnsafeBase base err val = Value val | Error err | Other (base val) deriving (Show, Eq) class MagicMerge m1 m2 | m1 -> m2 where magicMerge :: m1 a -> m2 a instance MagicMerge (UnsafeBase (UnsafeBase base err) err) (UnsafeBase base err) where magicMerge ma = case ma of Value a -> Value a Error e -> Error e Other o -> o instance (MagicMerge (UnsafeBase base err2) dstBase) => MagicMerge (UnsafeBase (UnsafeBase base err1) err2) (UnsafeBase dstBase err1) where magicMerge (ma :: UnsafeBase (UnsafeBase base err1) err2 a) = case ma of Value a -> Value a Error e -> Other $ magicMerge (Error e :: UnsafeBase base err2 a) Other o -> case o of Value a -> Other $ magicMerge (Value a :: UnsafeBase base err2 a) Error e -> Error e Other o' -> Other $ magicMerge (Other o' :: UnsafeBase base err2 a) main = print "help me world!" }}} The code compiles and works ok in both GHC-7.6 and GHC-7.8. In such cases, GHC-7.8 should inform about lacking premises. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9267> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #10313: ApiAnnotations : strings in warnings do not return SourceText
by GHC 25 Jul '15

25 Jul '15

#10313: ApiAnnotations : strings in warnings do not return SourceText -------------------------------------+------------------------------------- Reporter: alanz | Owner: alanz Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.1 Keywords: | Operating System: Unknown/Multiple ApiAnnotations | Type of failure: None/Unknown Architecture: | Blocked By: Unknown/Multiple | Related Tickets: Test Case: | Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- The `strings` used in a `WARNING` pragma are captured via {{{ strings :: { Located ([AddAnn],[Located FastString]) } : STRING { sL1 $1 ([],[L (gl $1) (getSTRING $1)]) } .. }}} The `STRING` token has a method `getSTRINGs` that returns the original source text for a string. A warning of the form {{{#!hs {-# WARNING Logic , mkSolver , mkSimpleSolver , mkSolverForLogic , solverSetParams , solverPush , solverPop , solverReset , solverGetNumScopes , solverAssertCnstr , solverAssertAndTrack , solverCheck , solverCheckAndGetModel , solverGetReasonUnknown "New Z3 API support is still incomplete and fragile: \ \you may experience segmentation faults!" #-} }}} returns the concatenated warning string rather than the original source. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10313> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #8327: Cmm sinking does not eliminate dead code in loops
by GHC 25 Jul '15

25 Jul '15

#8327: Cmm sinking does not eliminate dead code in loops ------------------------------------+------------------------------------- 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: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- Cmm code shown in #8326 is suboptimal. For this Haskell code: {{{ isTrue# :: Int# -> Bool isTrue# 1# = True isTrue# _ = False f :: Int# -> Int f x | isTrue# (x ># 0#) = I# x | otherwise = -(I# x) }}} we get Cmm that contains dead code in a loop: {{{ cFg: // False branch Hp = Hp + 16; if (Hp > HpLim) goto cFy; else goto cFx; cFy: // not enough heap, call GC HpAlloc = 16; I64[Sp - 16] = cFf; R1 = _sEV::I64; I64[Sp - 8] = _sEU::I64; Sp = Sp - 16; call stg_gc_unbx_r1(R1) returns to cFf, args: 8, res: 8, upd: 8; cFf: // re-do the False branch _sEU::I64 = I64[Sp + 8]; Sp = Sp + 16; _sEV::I64 = R1; goto cFg; cFx: // RHS of False branch I64[Hp - 8] = GHC.Types.I#_con_info; I64[Hp] = -_sEU::I64; R1 = Hp - 7; call (P64[Sp])(R1) args: 8, res: 0, upd: 8; }}} Notice that `_sEV` is dead - we should optimize it away. This is a known deficiency in the Cmm sinking pass (i.e. it does not work for loops). I'm putting this on Trac so we have this documented. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8327> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #8604: Some stack/vmem limits (ulimit) combinations causing GHC to fail
by GHC 24 Jul '15

24 Jul '15

#8604: Some stack/vmem limits (ulimit) combinations causing GHC to fail ----------------------------------+---------------------------------- Reporter: clavin | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.6.3 Keywords: | Operating System: Other Architecture: x86_64 (amd64) | Type of failure: Runtime crash Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ----------------------------------+---------------------------------- I have encountered a strange occurrence with GHC that was causing several GHC job failures on an SGE cluster. It turned out that there were other SGE users who needed an absurdly large stack size limit (set via 'ulimit -s') in the several gigabytes range. The default stack size limit had to be raised for the entire cluster. Any job that was run on a machine where the virtual memory limit was less than or equal to 2X the stack size limit, GHC would crash with the following message: ghc: failed to create OS thread: Cannot allocate memory I am running on GHC 7.6.3 with a 64-bit RedHat Enterprise OS, version 5.5. To reproduce the error, I was able to create the following test case: [ ~]$ ulimit -v unlimited [ ~]$ ulimit -s 10240 [ ~]$ ghc --version The Glorious Glasgow Haskell Compilation System, version 7.6.3 [ ~]$ ulimit -v 200000 [ ~]$ ulimit -s 100000 [ ~]$ ghc --version ghc: failed to create OS thread: Cannot allocate memory Several other programs work find using these settings, but GHC consistently has problems. Is this a fundamental issue with how GHC operates or can this be addressed? -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8604> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #9229: Compiler memory use regression
by GHC 24 Jul '15

24 Jul '15

#9229: Compiler memory use regression -------------------------+------------------------------------------------- Reporter: | Owner: augustss | Status: new Type: bug | Milestone: Priority: | Version: 7.8.2 normal | Operating System: Windows Component: | Type of failure: Compile-time performance bug Compiler | Test Case: Keywords: | Blocking: Architecture: | Unknown/Multiple | Difficulty: | Unknown | Blocked By: | Related Tickets: | -------------------------+------------------------------------------------- I have a module which compiles fine with ghc 7.6, but where ghc 7.8.2 runs out of memory (32-bit Windows). Sadly, I cannot supply the source code since it's proprietary, so I realize that this bug will be almost impossible to fix. It is possible to compile the module with -O0 and -O1, but not -O2. (This is a very performance critical module so I'd like -O2.) -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9229> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #9256: Support automatic derivation of an hs-boot file from an hs file
by GHC 24 Jul '15

24 Jul '15

#9256: Support automatic derivation of an hs-boot file from an hs file ------------------------------------+------------------------------------- Reporter: ezyang | Owner: ezyang Type: feature request | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.2 Keywords: backpack | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- This is essentially augustss's proposal from ticket #1409: I would settle for a solution where I annotate those entities in a module that would go in the hs-boot file. So instead of me having to go through the trouble of creating an extra file I could just say, e.g., Replying to [comment:13 augustss]: > I would settle for a solution where I annotate those entities in a module that would go in the hs-boot file. So instead of me having to go through the trouble of creating an extra file I could just say, e.g., > > {{{ > {-# MODULE_MUTUAL_RECURSION_BREAKER T, foo #-} > data T = ... > foo :: ... > foo = ... > }}} > > The values that have an annotation would have to have a type signature. This should be eminently doable. Here is a more fleshed out proposal. First, consider the definition of an hs-boot file in the absence of mutual recursion (i.e. no breakers are necessary). In this case, the hi-boot file for an hs file is precisely the hi file produced after ordinary renaming and typechecking of the module. As an optimization, we would like to avoid typechecking definitions if an explicit type signature is present (if the type signature is absent, we have no choice.) There are a few choices here: we can force users to give top-level declarations for all exported functions (so unconditionally error if something is missing, by filtering out the relevant value declarations from the source), or we can go ahead and do full typechecking when the signature is missing. (The former is probably the simplest.) Now, how to handle breakers? We need to avoid typechecking them, since their recursive dependence means that they would not type check properly. So we should also filter out their type signatures. (Furthermore, if we are inferring some signatures, we need to make sure that their definitions don't rely on any breakers. For this reason, it is probably simplest to just require explicit type signatures.) If the mechanism for specifying breakers is exclusionary (as opposed to an inclusionary method of saying what values are in the signature), there is an awkward problem of syntax for exporting only an abstract type: the logical choices mean the wrong thing (`T(..)` seem to imply the type is eliminated entirely, whereas `T` seems to imply that the type should be eliminated, but not eh constructors.) Summary: 1. Require top-level signatures in hs to hs-boot files(?) 2. Exclusionary or inclusionary signature pragma? 3. Filter out value declarations and excluded files, then rename/typecheck as normal, producing hi-boot file -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9256> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #10018: Cannot define custom fixity for infix data constructors in GHCi
by GHC 23 Jul '15

23 Jul '15

#10018: Cannot define custom fixity for infix data constructors in GHCi -------------------------------------+------------------------------------- Reporter: | Owner: RyanGlScott | Status: new Type: bug | Milestone: Priority: normal | Version: 7.8.4 Component: GHCi | Operating System: Unknown/Multiple Keywords: | Type of failure: Incorrect result Architecture: | at runtime Unknown/Multiple | Blocked By: Test Case: | Related Tickets: Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- When compiling from a module, declaring a custom fixity for an infix data constructor is valid: {{{#!hs module DataFixity where data Infix a b = a :@: b infixl 4 :@: }}} And GHCi recognizes this: {{{ $ ghci DataFixity.hs GHCi, version 7.8.4: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... done. [1 of 1] Compiling DataFixity ( DataFixity.hs, interpreted ) Ok, modules loaded: DataFixity. λ> :i :@: data Infix a b = a :@: b -- Defined at DataFixity.hs:3:18 infixl 4 :@: }}} However, one cannot do this entirely in GHCi: {{{ $ ghci GHCi, version 7.8.4: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... done. λ> data Infix a b = a :@: b; infixl 4 :@: λ> :i (:@:) data Infix a b = a :@: b -- Defined at <interactive>:2:18 }}} GHCi simply assumes {{{:@:}}} has the default infix precedence. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10018> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 8

[GHC] #10094: Template Haskell cannot represent type wildcards
by GHC 20 Jul '15

20 Jul '15

#10094: Template Haskell cannot represent type wildcards -------------------------------------+------------------------------------- Reporter: adamgundry | Owner: Type: feature | Status: new request | Milestone: Priority: normal | Version: 7.10.1-rc1 Component: Template | Operating System: Unknown/Multiple Haskell | Type of failure: GHC rejects Keywords: | valid program Architecture: | Blocked By: Unknown/Multiple | Related Tickets: Test Case: | Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- There does not appear to be a way for Template Haskell code to generate partial type signature wildcards. The type quasiquote `[t|_|]` results in `Wildcard not allowed`, and use in an expression quasiquote like `[|() :: _|]` results in `Exotic form of type not (yet) handled by Template Haskell _`. It would be useful if TH code could partially constrain the type of the expressions it was generating, without needing to determine the type completely. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10094> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #8131: T7571 with WAY=llvm fails, but not WAY=optllvm
by GHC 19 Jul '15

19 Jul '15

#8131: T7571 with WAY=llvm fails, but not WAY=optllvm ------------------------------------+------------------------------------- Reporter: thoughtpolice | Owner: Type: bug | Status: new Priority: high | 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: | ------------------------------------+------------------------------------- Test case (#7571 just will not die, it seems) {{{ $ make TEST=T7571 WAY=llvm ... ====> Scanning ./all.T =====> T7571(llvm) 6 of 7 [0, 0, 0] cd . && '/home/a/ghc/ghc-pristine/inplace/bin/ghc-stage2' -fforce-recomp -dcore-lint -dcmm-lint -dno-debug-output -no-user-package-db -rtsopts -fno-ghci-history -c T7571.cmm -fllvm -no-hs-main >T7571.comp.stderr 2>&1 Actual stderr output differs from expected: --- /dev/null 2013-08-07 22:18:14.132934017 -0500 +++ ./T7571.comp.stderr 2013-08-14 01:26:33.942103723 -0500 @@ -0,0 +1 @@ +WARNING: Non constant alignment value given for memcpy! Please report to GHC developers *** unexpected failure for T7571(llvm) }}} However, this test case does not fail under WAY=optllvm, which I think is worth investigating probably: {{{ $ make TEST=T7571 WAY=optllvm ... ====> Scanning ./all.T =====> T7571(optllvm) 6 of 7 [0, 0, 0] cd . && '/home/a/ghc/ghc-pristine/inplace/bin/ghc-stage2' -fforce-recomp -dcore-lint -dcmm-lint -dno-debug-output -no-user-package-db -rtsopts -fno-ghci-history -c T7571.cmm -O -fllvm -no-hs-main >T7571.comp.stderr 2>&1 OVERALL SUMMARY for test run started at Wed Aug 14 01:25:42 CDT 2013 7 total tests, which gave rise to 28 test cases, of which 27 were skipped 0 had missing libraries 1 expected passes 0 expected failures 0 caused framework failures 0 unexpected passes 0 unexpected failures }}} However, there is another way to trigger this bug in *both* cases: {{{ #include "Cmm.h" testMemcpy (W_ dst, W_ src, W_ l, W_ sz) { prim %memcpy(dst, src, l, sz); return (); } }}} will fail both ways. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8131> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 17

[GHC] #9571: nofib should use criterion-style bootstrapping/sampling
by GHC 19 Jul '15

19 Jul '15

#9571: nofib should use criterion-style bootstrapping/sampling -------------------------------------+------------------------------------- Reporter: ezyang | Owner: Type: feature request | Status: new Priority: normal | Milestone: Component: NoFib benchmark | Version: 7.9 suite | Operating System: Keywords: | Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: Difficulty: Unknown | None/Unknown Blocked By: | Test Case: Related Tickets: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- As I discovered when investigating situations like #9570, in some cases, test-cases in nofib are giving nonsense, and it's hard to tell unless you run nofib several times and notice that percentage differences are fluctuating up and down. The quality of the numbers we get for uninformed users would be better if we ran some statistical analysis to tell how many times to run the benchmark, and if there were lots of outliers (rather than just blindly summarizing all the runs using an average.) -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9571> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3