June 2016 - ghc-tickets

[GHC] #8949: switch -msse2 to be on by default
by GHC 21 Dec '18

21 Dec '18

#8949: switch -msse2 to be on by default ----------------------------------------------+---------------------------- Reporter: errge | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler (CodeGen) | Version: 7.9 Keywords: | Operating System: Architecture: x86 | Unknown/Multiple Difficulty: Moderate (less than a day) | Type of failure: Other Blocked By: | Test Case: Related Tickets: | Blocking: ----------------------------------------------+---------------------------- I propose msse2 to be on by default. I guess the default was chosen way back, when Pentium III support was still relevant. Nowadays we don't really win on the CPU support, because e.g. https://github.com/tibbe/hashable/blob/master/hashable.cabal is built by default with sse2 on the injected C code level. And hashable has a lot of reverse depends, therefore on an end user system (RedHat or Debian) the user is most probably unlucky with a Pentium III CPU anyways. Flipping this default would also fix the excess precision problem for most users of GHC on the i686 platform. GHC should provide a -mno-sse2 flag for the cases when this needs to be disabled. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8949> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 17

[GHC] #8655: Evaluate know-to-terminate-soon thunks
by GHC 19 Dec '18

19 Dec '18

#8655: Evaluate know-to-terminate-soon thunks ------------------------------------+------------------------------------- Reporter: nomeata | Owner: nomeata Type: task | 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: | ------------------------------------+------------------------------------- I guess I’ll better put my interior monologue in a ticket than on ghc- dev... In order to implement nested CPR (#1600), I had to enrich the CPR type to keep track of whether something, when entered, will converge for sure. My code does not solve the halting problem, but does only simple stuff, so if something is known to converge, is is very likely to be cheap. Simon suggested to measure the effect of evaluating a let-bound thunk with the known-to-terminate property, as if the body was using it strictly. This tickets tracks this suggestion, and reports progress. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8655> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 12

[GHC] #8598: IO hack in demand analyzer gets in the way of CPR
by GHC 19 Dec '18

19 Dec '18

#8598: IO hack in demand analyzer gets in the way of CPR ------------------------------------+------------------------------------- Reporter: nomeata | Owner: Type: task | 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: | ------------------------------------+------------------------------------- After a lot of staring at code and comparing unexpected nofib results I found the following: The IO hack in the demand analyzer (see `dmdAnalAlt` in `StrAnal.lhs` and #1592 for a good discussion) prevents CPR in any function that uses a C call. This is a small example, reduced from the `scaleFloat` method for doubles: {{{ module Float(fun) where import GHC.Float (Double(..)) import GHC.Integer (decodeDoubleInteger, encodeDoubleInteger) fun :: Double -> Double fun x | isFix = x | otherwise = case x of (D# x#) -> case decodeDoubleInteger x# of (# i, j #) -> D# (encodeDoubleInteger i j) where isFix = isDoubleFinite x == 0 foreign import ccall unsafe "isDoubleFinite" isDoubleFinite :: Double -> Int }}} Here, `fun` does current not get the CPR property, and the work gets type `GHC.Prim.Double# -> GHC.Types.Double`. Why? Because in core, there will be a {{{ case {__pkg_ccall main isDoubleFinite GHC.Prim.Double# -> GHC.Prim.State# GHC.Prim.RealWorld -> (# GHC.Prim.State# GHC.Prim.RealWorld, GHC.Prim.Int# #)}_dQD ds_dQA GHC.Prim.realWorld# of _ [Occ=Dead, Dmd=<L,A>] { (# ds_dQC [Dmd=<L,A>, OS=OneShot], ds_dQB [Dmd=<S,1*U>] #) -> ... }}} where the case body has `DmdType m {dQz-><L,1*U(U)> dQA-><L,U>}`, but `dmdAnalAlt` detects that this is evaluating a possibly exiting function and throws aways the information using `alt_ty `lubDmdType` topDmdType`. Would it be ok to `lub` only the demand on the free variables, but keep the CPR information? In nofib (if I managed to compare the right results) this does nothing for almost all benchmarks, `-9.2%` of allocations for `mandel` and `+4.9%` for reverse-complement (but these numbers are not obtained very cleanly, and all is happening on top of the better-ho-cardinality branch. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8598> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 15

[GHC] #10905: Incorrect number of parameters in "role" errors
by GHC 19 Dec '18

19 Dec '18

#10905: Incorrect number of parameters in "role" errors -------------------------------------+------------------------------------- Reporter: crockeea | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.2 Keywords: roles | Operating System: Unknown/Multiple Architecture: | Type of failure: Incorrect Unknown/Multiple | warning at compile-time Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- The following code produces the error: {{{ import Data.Coerce import Data.Functor.Trans.Tagged toTT :: Tagged t [a] -> TaggedT t [] a toTT = coerce }}} {{{ Couldn't match representation of type ‘[a]’ with that of ‘Data.Functor.Identity.Identity [a]’ arising from trying to show that the representations of ‘Tagged t [a]’ and ‘TaggedT t [] a’ are the same Relevant role signatures: type role [] representational type role Data.Functor.Identity.Identity representational type role TaggedT nominal nominal phantom representational nominal The data constructor ‘Data.Functor.Identity.Identity’ of newtype ‘Data.Functor.Identity.Identity’ is not in scope Relevant bindings include toTT :: Tagged t [a] -> TaggedT t [] a (bound at Main.hs:9:1) In the expression: coerce In an equation for ‘toTT’: toTT = coerce }}} First, kudos for correctly identifying the problem: I failed to import the `Identity` constructor. However, it seems that the report for the roles of the `TaggedT` type has too many parameters. It is defined in tagged-transformer [https://hackage.haskell.org/package/tagged-transformer-0.8/docs/Data- Functor-Trans-Tagged.html] as `newtype TaggedT s m b = TagT { untagT :: m b }`, but is listed with 5 roles. I did try to move the definition of `TaggedT/Tagged` into my module so I didn't have to import the library, but when I did the error didn't show up: TaggedT was listed with the expected roles `phantom representational nominal`. As a user, it wasn't clear to me why there are 5 roles listed for a type with 3 parameters, and I didn't know which roles I should be looking at to help me debug the `coerce` error. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10905> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #8959: GHCi should honour UnicodeSyntax
by GHC 19 Dec '18

19 Dec '18

#8959: GHCi should honour UnicodeSyntax -------------------------------------------+------------------------------- Reporter: Fuuzetsu | Owner: Type: bug | Status: new Priority: low | Milestone: Component: Compiler | Version: 7.6.3 Keywords: | Operating System: Architecture: Unknown/Multiple | Unknown/Multiple Difficulty: Easy (less than 1 hour) | Type of failure: Blocked By: | None/Unknown Related Tickets: | Test Case: | Blocking: -------------------------------------------+------------------------------- When we load a module into GHCi with -XUnicodeSyntax, GHCi should default to printing information with the relevant unicode symbols. {{{ *Communication> let a ∷ Int; a = 7 *Communication> :t a a :: Int }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8959> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 27

[GHC] #11042: Template Haskell / GHCi does not respect extra-lib-dirs
by GHC 11 Dec '18

11 Dec '18

#11042: Template Haskell / GHCi does not respect extra-lib-dirs -------------------------------------+------------------------------------- Reporter: mboes | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.2 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: #10458 #5289 Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- In particular when compiling on NixOS, but also in many other situations, it is often needed to pass --extra-lib-dirs / --extra-include-dirs to cabal/stack, to help GHC find required system libraries. This works fine for GHC, but not GHCi, and therefore also not Template Haskell code. Minimal example attached. Try it in an environment where libz is not in a standard location, e.g. inside a nix-shell. Note that if you remove the template Haskell splice in Main.hs, the code compiles just fine. This affects GHC 7.10.2, but also HEAD from 2015-08-28, even one including Tamar Christina's 5d841108acef950fed6a5e608ac9b18e7431aa87 commit cherry- picked. I haven't been able to test more recent HEAD, because of unrelated compilation issues. I surmise that a fix to #10458, which currently blocks using HaskellR in GHCi, would also fix this, but I'm not sure. The error message I get is different: {{{ $ stack build --skip-ghc-check --extra-lib- dirs=/nix/store/31w31mc8immhpnmxvcl4l0fvc3i5iwh0-zlib-1.2.8/lib --extra- include- dirs=/nix/store/31w31mc8immhpnmxvcl4l0fvc3i5iwh0-zlib-1.2.8/include b-0.1.0.0: build Preprocessing executable 'b' for b-0.1.0.0... [2 of 2] Compiling Main ( src/Main.hs, .stack- work/dist/x86_64-linux/Cabal-1.23.0.0/build/b/b-tmp/Main.o ) <command line>: can't load .so/.DLL for: libz.so (libz.so: cannot open shared object file: No such file or directory) -- While building package b-0.1.0.0 using: /home/mboes/.stack/setup-exe-cache/setup-Simple- Cabal-1.23.0.0-x86_64-linux-ghc-7.11.20150828 --builddir=.stack- work/dist/x86_64-linux/Cabal-1.23.0.0/ build exe:b --ghc-options " -ddump- hi -ddump-to-file" Process exited with code: ExitFailure 1 }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11042> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 38

[GHC] #9718: Avoid TidyPgm predicting what CorePrep will do
by GHC 10 Dec '18

10 Dec '18

#9718: Avoid TidyPgm predicting what CorePrep will do -------------------------------------+------------------------------------- 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: -------------------------------------+------------------------------------- At the moment the `TidyPgm` pass is forced to predict, accurately but unpleasantly, some aspects of what `CorePrep` and `Core2Stg` will do. Reason: * Each `Id` in the interface file records (among other things) the '''arity''' of the `Id`, and whether it has '''CAF references''' * We really only know these two things for sure after `CorePrep`. The conversion from Core to STG makes no structural changes. * However the result of `TidyPgm` (which preceded `CorePrep`) is used to generate the interface file. So it has to predict the arity and CAF-ref status of each `Id`. * This is not good. It restricts what `CorePrep` can do (notably, it must not change the arity of a top-level `Id`), and it leads to unsavoury code (e.g. look at the call to `CorePrep.cvtLitInteger` in `TidyPgm.cafRefsL`. It's also dangerous: an inconsistency could lead to a crash. This is a long-standing problem. My current thought for how to unravel it is this: * `TidyPgm` does not attach arity or CAF-ref info. * Instead, run `CorePrep` after `TidyPgm`, and generate accurate arity and CAF-ref info * Then use that auxiliary mapping during the conversion from tidied program to `ModIface`. I don't think this would be hard. It would mean that the tidied program and the core-prep'd program would have to exist in memory at the same time. An alternative would be to generate the `ModIface` from the tidied program ''sans'' arity and CAF-ref info, and then, after `CorePrep` run over it to add arity and CAF-ref info. (You'd have to do this before generating the fingerprints.) The advantage of this is that the `ModIface` can be a lot smaller than the code for the entire module. A long-standing wart which needs some careful attention. Simon -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9718> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 42

[GHC] #8109: Type family patterns should support as-patterns.
by GHC 09 Dec '18

09 Dec '18

#8109: Type family patterns should support as-patterns. ------------------------------------+------------------------------------- Reporter: carlhowells | 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: | ------------------------------------+------------------------------------- I recently wrote code similar to the following: {{{ {-# LANGUAGE DataKinds, ScopedTypeVariables, TypeOperators, TypeFamilies #-} {-# LANGUAGE PolyKinds, FlexibleInstances, FlexibleContexts #-} import GHC.TypeLits data P n = P fromNat :: forall (p :: Nat -> *) (n :: Nat). SingI n => p n -> Integer fromNat _ = fromSing (sing :: Sing n) class C (a :: [Nat]) where type T a :: * val :: p a -> T a instance SingI n => C '[n] where type T '[n] = Integer val _ = fromNat (P :: P n) instance (SingI n, C (n2 ': ns)) => C (n ': n2 ': ns) where type T (n ': n2 ': ns) = (Integer, T (n2 ': ns)) val _ = (fromNat (P :: P n), val (P :: P (n2 ': ns))) }}} There were semantic constraints in my case that made an empty list nonsensical, so it wasn't an appropriate base case. But the resulting effort in ensuring type family patterns didn't overlap got unwieldy. I would have much preferred to write that second instance more like this: {{{ instance (SingI n, C ns) => C (n ': ns@(_ ': _)) where type T (n ': ns) = (Integer, T ns) val _ = (fromNat (P :: P n), val (P :: P ns)) }}} The reasoning here is identical to as-patterns in value-level pattern matching. If you only want to match an expression when it has a particular sub-structure, it's way more convenient to do it with an as-pattern. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8109> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 8

[GHC] #12126: Bad error messages for SPECIALIZE pragmas
by GHC 09 Dec '18

09 Dec '18

#12126: Bad error messages for SPECIALIZE pragmas -------------------------------------+------------------------------------- Reporter: antalsz | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 Keywords: Specialize, | Operating System: Unknown/Multiple ErrorMessages | Architecture: | Type of failure: Incorrect Unknown/Multiple | warning at compile-time Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- When playing with SPECIALIZE pragmas, I've encountered one large problem with the error messages, and two minor ones. 1. Given a function with a `Num` constraint, applying `SPECIALIZE` at a non-function type triggers defaulting incorrectly; the code {{{#!hs f1 :: Num a => a -> a f1 x = x + x {-# SPECIALIZE f1 :: Bool #-} }}} produces the error {{{ …/Test1.hs:5:1: error: • Couldn't match expected type ‘Integer -> Integer’ with actual type ‘Bool’ • In the SPECIALISE pragma {-# SPECIALIZE f1 :: Bool #-} }}} (This is in the attached file `Test1.hs`.) 2. Given a function, applying `SPECIALIZE` at a well-formed type without the looked-for instance produces an error with some extra text (the first line ends with "a SPECIALISE pragma for ‘f2’", even though this is repeated on the next line); the code {{{#!hs f2 :: Num a => a -> a f2 x = x + x {-# SPECIALIZE f2 :: Bool -> Bool #-} }}} produces the error {{{ …/Test2.hs:5:1: error: • No instance for (Num Bool) a SPECIALISE pragma for ‘f2’ • In the SPECIALISE pragma {-# SPECIALIZE f2 :: Bool -> Bool #-} }}} (This is in the attached file `Test2.hs`.) 3. This is minor, but: both of these error messages talk about a `SPECIALISE` pragma when I wrote a `SPECIALIZE` pragma. Even more so, case is not preserved in the quoted pragma; if I provide a {{{#!hs {-# specialize … #-} }}} pragma, the error message ends with {{{ • In the SPECIALISE pragma {-# SPECIALIZE … #-}`. }}} even though that isn't what I wrote. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12126> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 8

[GHC] #12088: Promote data family instance constructors
by GHC 06 Dec '18

06 Dec '18

#12088: Promote data family instance constructors -------------------------------------+------------------------------------- Reporter: alexvieth | Owner: Type: feature | Status: new request | Priority: normal | Milestone: Component: Compiler | Version: 8.1 (Type checker) | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: GHC rejects Unknown/Multiple | valid program Test Case: | Blocked By: Blocking: | Related Tickets: #11348 Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- In `TcEnv.hs` there is a note AFamDataCon: not promoting data family constructors. It states that we can't use a promoted data family instance constructor because we would have to interleave the checking of instances and data types. But with the fix of #11348, we now do exactly this. In the example from the note {{{#!hs data family T a data instance T Int = MkT data Proxy (a :: k) data S = MkS (Proxy 'MkT) }}} -ddump-rn-trace shows these groups {{{ rnTycl dependency analysis made groups [[data family T a_apG] [] [data instance T Int = MkT], [data Proxy (a_apF :: k_apE)] [] [], [data S = MkS (Proxy MkT)] [] []] }}} That's to say, the instance `T Int` will in fact be checked before `S`. So let's remove this restriction. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12088> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 53