December 2014 - ghc-tickets

[GHC] #8287: exploring calling convention changes and related engineering for 7.10
by GHC 25 Jan '16

25 Jan '16

#8287: exploring calling convention changes and related engineering for 7.10 ------------------------------------+------------------------------------- Reporter: carter | Owner: Type: task | Status: new Priority: high | Milestone: 7.10.1 Component: Compiler | Version: 7.6.3 Keywords: | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: None/Unknown Difficulty: Rocket Science | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- I'm creating this as a master ticket for systematically exploring and benchmarking (possibly breaking) changes to the GHC calling convention. This work will initially focus on x86/x86_64, but may have larger scope. Goals: 1. if possible, improve performance systematically for code run on recent CPU micro architecture generations (eg sandybridge, haswell, and future extensions like Knights Landing and Skylake), ideally without pessimizing code on other or older x86_64 micro-architectures 2. Try to bring native and llvm codegens closer to feature parity (or at the very least, do not widen their capability gap) 3. a few other pieces too, will amend this ticket as ideas / plans clarify -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8287> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 11

[GHC] #9149: Description of openFileBlocking is wrong
by GHC 25 Jan '16

25 Jan '16

#9149: Description of openFileBlocking is wrong ------------------------------------+------------------------------------- Reporter: edsko | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Documentation | 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: | ------------------------------------+------------------------------------- The documentation states: ''Like openFile, but opens the file in ordinary blocking mode. This can be useful for opening a FIFO for reading: if we open in non-blocking mode then the open will fail if there are no writers, whereas a blocking open will block until a writer appears. '' but I think this is the wrong way around, and should be: ''Like openFile, but opens the file in ordinary blocking mode. This can be useful for opening a FIFO for writing: if we open in non-blocking mode then the open will fail if there are no readers, whereas a blocking open will block until a reader appears. '' -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9149> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #8325: Pattern guards in anonymous functions
by GHC 25 Jan '16

25 Jan '16

#8325: Pattern guards in anonymous functions ------------------------------------+------------------------------------- Reporter: mcollis | 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: | ------------------------------------+------------------------------------- In the course of a project, I had an idea that would allow less cumbersome use of cases in anonymous functions. Here's a simple example of what I'm proposing: {{{ (\x y | x > y -> x | otherwise -> something else) }}} Currently, the only way to accomplish this is with if() then ... else syntax or the use of the MultiWayIf and LambdaCase extensions. The syntax I've proposed here matches with existing anonymous function syntax and doesn't conflict with any other GHC syntax. This would greatly improve the readability of complex anonymous functions. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8325> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #9708: Type inference non-determinism due to improvement
by GHC 24 Jan '16

24 Jan '16

#9708: Type inference non-determinism due to improvement -------------------------------------+------------------------------------- Reporter: simonpj | 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: -------------------------------------+------------------------------------- Here's an example that showed up in Iavor's test `TcTypeNatSimple`: {{{ ti7 :: (x <= y, y <= x) => Proxy (SomeFun x) -> Proxy y -> () ti7 _ _ = () }}} From the ambiguity check for `ti7` we get this constraint {{{ forall x y. (x <= y, y <= x) => ( SomeFun x ~ SomeFun alpha , alpha <= y, y <= alpha) }}} where `alpha` is the unification variable that instantiates `x`. Now, from the givens, improvement gives a derived `[D] x~y`, but we can't actually use that to rewrite anything; we have no evidence. From the wanteds we can use improvement in two alternative ways: {{{ 1. (alpha <= y, y <= alpha) => [D] alpha ~ y 2. (x <= y, y <= alpha) => [D] x <= alpha (alpha <= y, y <= x) => [D] alpha <= x And combining the latter two we get [D] (alpha ~ x) }}} It is (2) that we want. But if we get (1) and fix `alpha := y`, we get an error `Can't unify (SomeFun x ~ SomeFun y)`. I think it's a fluke that this test has been working so far; in my new flatten-skolem work it has started failing, which is not unreasonable. What I HATE is that whether type checking succeeds or fails depends on some random choice about which constraint is processed first. The real bug is that the derived Given has no evidence, and can't be used for rewriting. I think Iavor is fixing this deficiency for. I speculate that the same problem might also show up with ordinary type-class functional dependencies, but I have not found a concrete example. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9708> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #8572: Building an empty module with profiling requires profiling libraries for integer-gmp
by GHC 24 Jan '16

24 Jan '16

#8572: Building an empty module with profiling requires profiling libraries for integer-gmp -------------------------------------------+------------------------------- Reporter: parcs | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler (Type checker) | Version: 7.7 Keywords: | Operating System: Architecture: Unknown/Multiple | Unknown/Multiple Difficulty: Unknown | Type of failure: Blocked By: | None/Unknown Related Tickets: | Test Case: | Blocking: -------------------------------------------+------------------------------- {{{ #!haskell {-# LANGUAGE NoImplicitPrelude #-} module A where }}} {{{ $ ghc-stage2 -prof -c A.hs Top level: Failed to load interface for ‛GHC.Integer.Type’ Perhaps you haven't installed the profiling libraries for package ‛integer-gmp’? Use -v to see a list of the files searched for. }}} I can't built module `A` without profiling libraries for `integer-gmp`, even though I don't use `integer-gmp` anywhere in the module. This happens because the `Tidy Core` pass attempts to look up the `mkInteger` name (in order to desugar integer literals) even when there are no integer literals in the module. The obvious fix is to lazily look up `mkInteger` in `Coreprep.lookupMkIntegerName`: {{{ #!diff diff --git a/compiler/coreSyn/CorePrep.lhs b/compiler/coreSyn/CorePrep.lhs index 5e0cd65..9836982 100644 --- a/compiler/coreSyn/CorePrep.lhs +++ b/compiler/coreSyn/CorePrep.lhs @@ -56,6 +56,7 @@ import Config import Data.Bits import Data.List ( mapAccumL ) import Control.Monad +import System.IO.Unsafe ( unsafeInterleaveIO ) \end{code} -- --------------------------------------------------------------------------- @@ -1119,6 +1120,7 @@ lookupMkIntegerName dflags hsc_env else if thisPackage dflags == integerPackageId then return $ panic "Can't use Integer in integer" else liftM tyThingId + $ unsafeInterleaveIO $ initTcForLookup hsc_env (tcLookupGlobal mkIntegerName) mkInitialCorePrepEnv :: DynFlags -> HscEnv -> IO CorePrepEnv }}} This way, we don't attempt to look up `mkInteger` until we actually need it, i.e. if there are integer literals that we must desugar. Relevant commits are 2ef5cd26db27543ac8664a3d18f45550d0109a8b and fdd552e0ecaa17300670a48562995040e1d6687e -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8572> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #8285: unexpected behavior with encodeFloat on large inputs
by GHC 23 Jan '16

23 Jan '16

#8285: unexpected behavior with encodeFloat on large inputs --------------------------+------------------------------------------------ Reporter: carter | Owner: Type: bug | Status: new Priority: normal | Milestone: 7.8.1 Component: | Version: 7.7 Compiler | Operating System: Unknown/Multiple Keywords: | Type of failure: Incorrect result at runtime Architecture: | Test Case: Unknown/Multiple | Blocking: Difficulty: | Unknown | Blocked By: | Related Tickets: | --------------------------+------------------------------------------------ benjamin scarlet discovered some unexpected behavior in encodeFloat today {{{ <bscarlet> > map (\x -> encodeFloat 1 (shiftL 1 x)) [30,31] [23:47:47] <lambdabot> [Infinity,0.0] [23:49:09] <arkeet> > encodeFloat 1 60 [23:49:10] <lambdabot> 1.152921504606847e18 [23:49:09] <arkeet> > encodeFloat 1 60 [23:49:10] <lambdabot> 1.152921504606847e18 [23:49:18] <arkeet> er. [23:50:21] <bscarlet> down in the sane range I suspect it's just fine. [23:50:43] <bscarlet> > encodeFloat 1 1000 [23:50:44] <lambdabot> 1.0715086071862673e301 [23:50:52] <bscarlet> > encodeFloat 1 10000 [23:50:53] <lambdabot> Infinity [23:50:58] <bscarlet> That's sane. [23:51:03] <arkeet> > encodeFloat 1 (2^31) [23:51:04] <lambdabot> 0.0 [23:51:13] <bscarlet> That's what I think is weird. [23:51:19] <carter> bscarlet floats are weird [23:51:23] <arkeet> > encodeFloat 1 (2^31-1) [23:51:24] <lambdabot> Infinity }}} Benjamin did a wee bit of investigation, and the problem seems to lie in https://github.com/ghc/packages-integer- gmp/blob/master/cbits/float.c#L177 in integer-gmp, where it deliberately doesn't handle the edge cases (probably worth investigating) if integer-simple has similar behavior or not -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8285> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #8403: Pretty-printing of long types
by GHC 23 Jan '16

23 Jan '16

#8403: Pretty-printing of long types ------------------------------------+------------------------------------- Reporter: monoidal | Owner: Type: bug | Status: new Priority: lowest | Milestone: Component: Compiler | Version: 7.6.3 Keywords: | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: Other Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- Low priority, but this ghci output could be more packed. {{{ λ> :t (,,,,,,,,,,,,,,,,,,,,,,,) (,,,,,,,,,,,,,,,,,,,,,,,) :: a -> b -> c -> d -> e -> f -> g -> h -> i -> j -> k -> l -> m -> n -> o -> p -> q -> r -> s -> t -> u -> v -> w -> x -> (a, b, c, d, e, f, g, h, i, j, k, l, m, n, o, p, q, r, s, t, u, v, w, x) }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8403> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #8064: Warning out when "undefined" value has been used.
by GHC 23 Jan '16

23 Jan '16

#8064: Warning out when "undefined" value has been used. ------------------------------------+------------------------------------- Reporter: freizl | 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: | ------------------------------------+------------------------------------- Programmer may like to use "undefined" to define a function or value, which usually means "later implementation". However, people might forget to implement it thus cause run time error. Wish GHC will show some warning/errors at compilation. This idea comes from this article (http://www.devalot.com/articles/2013/07 /why-haskell.html) -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8064> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

[GHC] #8422: type nats solver is too weak!
by GHC 23 Jan '16

23 Jan '16

#8422: type nats solver is too weak! -------------------------------------+------------------------------------- Reporter: carter | Owner: Type: bug | Status: new Priority: highest | Milestone: 7.8.1 Component: Compiler (Type | Version: 7.7 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: | -------------------------------------+------------------------------------- I just built ghc HEAD today, and the type nat solver can't handle the attached program, which *should* be simple to check! (and while I could use unsafeCoerce to "prove" it correct, that defeats the purpose of having the type nat solver!) -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8422> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #8618: can't load .so/.DLL
by GHC 22 Jan '16

22 Jan '16

#8618: can't load .so/.DLL -----------------------------------+--------------------------------------- Reporter: Feuerbach | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler (NCG) | Version: 7.7 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Compile-time crash Unknown/Multiple | Test Case: Difficulty: Unknown | Blocking: Blocked By: | Related Tickets: | -----------------------------------+--------------------------------------- This happens in a fresh sandbox: {{{ % cabal install uniplate -j Resolving dependencies... Configuring syb-0.4.1... Configuring text-1.0.0.0... Building syb-0.4.1... Building text-1.0.0.0... Installed syb-0.4.1 Installed text-1.0.0.0 Configuring hashable-1.2.1.0... Building hashable-1.2.1.0... Installed hashable-1.2.1.0 Configuring unordered-containers-0.2.3.3... Building unordered-containers-0.2.3.3... Installed unordered-containers-0.2.3.3 Configuring uniplate-1.6.12... Building uniplate-1.6.12... Failed to install uniplate-1.6.12 Last 10 lines of the build log ( /home/feuerbach/.cabal/logs/uniplate-1.6.12.log ): Preprocessing library uniplate-1.6.12... [ 1 of 21] Compiling Data.Generics.Uniplate.Data.Instances ( Data/Generics/Uniplate/Data/Instances.hs, dist/build/Data/Generics/Uniplate/Data/Instances.o ) [ 2 of 21] Compiling Data.Generics.Uniplate.Internal.Utils ( Data/Generics/Uniplate/Internal/Utils.hs, dist/build/Data/Generics/Uniplate/Internal/Utils.o ) Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... done. Loading package array-0.5.0.0 ... linking ... done. Loading package deepseq-1.3.0.2 ... linking ... done. Loading package bytestring-0.10.4.0 ... linking ... done. Loading package text-1.0.0.0 ... <command line>: can't load .so/.DLL for: libHStext-1.0.0.0.so (libHStext-1.0.0.0.so: cannot open shared object file: No such file or directory) cabal: Error: some packages failed to install: uniplate-1.6.12 failed during the building phase. The exception was: ExitFailure 1 }}} I have shared: False in cabal.config, so it's not surprising that libHStext-1.0.0.0.so is not found. What is surprising is that it's needed. uniplate doesn't use TemplateHaskell, from what I can tell. This is GHC 7.7.20131217. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8618> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 11