November 2013 - ghc-tickets

[GHC] #8095: TypeFamilies painfully slow
by GHC 08 Mar '19

08 Mar '19

#8095: TypeFamilies painfully slow -------------------------------------+------------------------------------- Reporter: MikeIzbicki | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler (Type | Version: 7.6.3 checker) | Operating System: Unknown/Multiple Keywords: | Type of failure: Compile-time Architecture: Unknown/Multiple | performance bug Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: 5321 | -------------------------------------+------------------------------------- I'm using the TypeFamilies extension to generate types that are quite large. GHC can handle these large types fine when they are created manually, but when type families get involved, GHC's performance dies. Unlike in ticket #5321, using tail recursion does not eliminate the problem, and the order of arguments greatly affects compile time. I've attached a file Types.hs that demonstrates the problems. This file generates another Haskell file which has the problems. It takes 3 flags. The first is the size of the type to generate, the second is which type family function to use, and the third is whether to call the type family or just use a manually generated type. Here are my performance results: Using non-tail recursion, I get these results. I have to increase the stack size based on the size of the type I want to generate. {{{ $ ./Types 200 a a > test.hs && time ghc test.hs > /dev/null -fcontext- stack=250 real 0m2.973s $ ./Types 300 a a > test.hs && time ghc test.hs > /dev/null -fcontext- stack=350 real 0m6.018s $ ./Types 400 a a > test.hs && time ghc test.hs > /dev/null -fcontext- stack=450 real 0m9.995s $ ./Types 500 a a > test.hs && time ghc test.hs > /dev/null -fcontext- stack=550 real 0m15.645s }}} Tail recursion generates much slower compile times for some reason, and I still need to adjust the stack size: {{{ $ ./Types 200 b a > test.hs && time ghc test.hs > /dev/null -fcontext- stack=250 real 0m16.120s }}} Changing the order of arguments to the recursive type family greatly changes the run times: {{{ $ ./Types 200 c a > test.hs && time ghc test.hs > /dev/null -fcontext- stack=250 real 0m6.095s }}} Without the type family, I get MUCH better performance: {{{ $ ./Types 10000 a d > test.hs && time ghc test.hs > /dev/null real 0m2.271s }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8095> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 64

[GHC] #8326: Place heap checks common in case alternatives before the case
by GHC 07 Jan '19

07 Jan '19

#8326: Place heap checks common in case alternatives before the case ------------------------------------+------------------------------------- Reporter: jstolarek | Owner: Type: task | 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: 8317 Related Tickets: #1498 | ------------------------------------+------------------------------------- We would like to have functions that check whether an `Int#` is a valid tag to represent `Bool` (see Note [Optimizing isTrue#] in ghc-prim): {{{ isTrue# :: Int# -> Bool isTrue# 1# = True isTrue# _ = False isFalse# :: Int# -> Bool isFalse# 0# = True isFalse# _ = False }}} We could use them with comparison primops like this: {{{ f :: Int# -> Int f x | isTrue# (x ># 0#) = I# x | otherwise = -(I# x) }}} `isTrue#` is optimized away at the Core level: {{{ A.f = \ (x_aqM :: GHC.Prim.Int#) -> case GHC.Prim.># x_aqM 0 of _ { __DEFAULT -> GHC.Types.I# (GHC.Prim.negateInt# x_aqM); 1 -> GHC.Types.I# x_aqM } }}} but the code genrator produces very bad Cmm code, because it pushes heap checks into case alternatives: {{{ {offset cFd: // stack check if ((Sp + -16) < SpLim) goto cFr; else goto cFs; cFr: // not enough place on the stack, call GC R2 = R2; R1 = A.f_closure; call (stg_gc_fun)(R2, R1) args: 8, res: 0, upd: 8; cFs: // scrutinize (x ># 0#) _sEU::I64 = R2; _sEV::I64 = %MO_S_Gt_W64(R2, 0); if (_sEV::I64 != 1) goto cFg; else goto cFo; 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; cFo: // True branch Hp = Hp + 16; if (Hp > HpLim) goto cFv; else goto cFu; cFv: // not enough heap, call GC HpAlloc = 16; I64[Sp - 16] = cFn; R1 = _sEV::I64; I64[Sp - 8] = _sEU::I64; Sp = Sp - 16; call stg_gc_unbx_r1(R1) returns to cFn, args: 8, res: 8, upd: 8; cFn: // re-do the True branch _sEU::I64 = I64[Sp + 8]; Sp = Sp + 16; _sEV::I64 = R1; goto cFo; cFu: // RHS of True 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; } }}} This results in average 2.5% increase in binary size. By contrast, if we use `tagToEnum#` instead of `isTrue#` heap check will be placed before `case` expression and the code will be significantly shorter (this is done by a special case-on-bool optimization in the code generator - see #8317). What we would like to do here is: 1. compile case alternatives without placing heap checks inside them 2. each compiled alternative should return amount of heap it needs to allocate 3. code generator inspects amounts of heap needed by each alternative and either adds heap checks in alternatives or puts a single check before the case expression. Getting this right might be a bit tricky. 1. if all branches allocate some heap then we can just put a common heap check before the case. Note that we must allocate the higgest amount required by any of the alternatives and then alternatives that use less heap must retract the heap pointer accordingly. 2. if we have two alternatives, one of which allocates heap and the other does not, we should place the heap check only in the alternative that allocates the stack. This will solve #1498. 3. it is not clear to me what to do if we have combination of the above (more than one branch that allocates heap and at least one branch that does not). If we place heap check before the `case` expression we lose optimization of recursive functions and face the problem described in #1498. If we push heap checks into branches that allocate heap then we get code duplication, i.e. the problem that we're addressing in this ticket. I guess the only way to make correct decission here is to try different aproaches and measure their performance. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8326> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 48

[GHC] #8316: GHCi debugger segfaults when trying force a certain variable
by GHC 28 Dec '18

28 Dec '18

#8316: GHCi debugger segfaults when trying force a certain variable ------------------------------------+------------------------------------- Reporter: guest | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: GHCi | Version: 7.6.3 Keywords: | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: GHCi crash Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- The file Test.hs has following definitions: {{{ whnf :: a -> IO () whnf a = a `seq` (return ()) foo :: [Int] foo = [1..] }}} Calling ghci as: {{{ ghci Test.hs -ignore-dot-ghci }}} and bebugging foo like this: {{{ *Main> :b foo Breakpoint 0 activated at Test.hs:5:7-11 *Main> foo Stopped at Test.hs:5:7-11 _result :: [Int] = _ [Test.hs:5:7-11] *Main> :p foo foo = (_t1::[Int]) [Test.hs:5:7-11] *Main> whnf _t1 }}} results in this segault: {{{ <interactive>: internal error: TSO object entered! (GHC version 7.6.3 for x86_64_unknown_linux) Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug [1] 5445 abort (core dumped) ghci Test.hs -ignore-dot-ghci }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8316> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 25

[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] #8043: Feature Request : Qualified module exports
by GHC 01 Dec '18

01 Dec '18

#8043: Feature Request : Qualified module exports ------------------------------------+------------------------------------- Reporter: erikd | 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 quite like the practice of using `-XNoImplicitPrelude` and then defining my own (often per-project) Prelude doing things like: {{{ module MyProject.Prelude ( module P ) import Data.Text as P (Text) import Prelude as P hiding (head, init, tail, last) }}} As an extension to the above idea, I would like to add qualified module exports defined so that I can do this: {{{ module MyProject.Prelude ( module P , qualified module T ) import Data.Text as P (Text) import Prelude as P hiding (head, init, tail, last) import qualified Data.Text as T }}} so that any file which imports `MyProject.Prelude` will then have all the identifiers that they would normally get if they did: {{{ import qualified Data.Text as T }}} Is this something that would have to go through the Haskell Next procedure? If so, how does one get that kicked off? -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8043> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 13

[GHC] #8177: Roles for type families
by GHC 29 Nov '18

29 Nov '18

#8177: Roles for type families ------------------------------------+------------------------------------- Reporter: simonpj | 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: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- Now that we have [wiki:Roles roles], it might be helpful to be able to give a role signature for a data/type family. At the moment, data/type family constructors have all parameters conservatively assigned to be role `N`. Thus {{{ data family D a -- Parameter conservatively assumed to be N class C a where op :: D a -> a instance C Int where .... newtype N a = MkN a deriving( C ) -- Rejected }}} The generalised-newtype-deriving clause `deriving( C )` is rejected because `D` might use its parameter at role `N` thus: {{{ data instance D [b] = MkD (F b) -- F is a type function }}} It would be strictly more expressive if we could * '''Declare''' the roles of D's arguments (as we can declare their kinds). E.g. {{{ data family D a@R }}} * '''Check''' that each family instance obeys that role signature. E.g. given the preceding role signature, reject this instance: {{{ data instance D [b] = MkD (F b) -- F is a type function }}} I think there is no technical difficulty here. Just a question of doing it. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8177> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 41

[GHC] #8226: Remove the old style -- # Haddock comments.
by GHC 01 Oct '18

01 Oct '18

#8226: Remove the old style -- # Haddock comments. ------------------------------------+------------------------------------- Reporter: Fuuzetsu | Owner: Type: task | Status: new Priority: normal | Milestone: 7.8.1 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: | ------------------------------------+------------------------------------- Haddock 0.x supported -- # style comments for Haddock pragma and while this seems to have been long forgotten and not documented, it's still kicking in the lexer. I think this is a good time to get rid of these all together. Relevant Haddock ticket http://trac.haskell.org/haddock/ticket/171 I'll hopefully send in patches removing these from the lexer/parser tomorrow, if all goes well (unless anyone feels eager to do it, then please, you're most welcome to do so!). -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8226> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 11

[GHC] #8400: Migrate the RTS to use libuv (or libev, or libevent)
by GHC 29 Sep '18

29 Sep '18

#8400: Migrate the RTS to use libuv (or libev, or libevent) ----------------------------------------------+---------------------------- Reporter: schyler | Owner: Type: feature request | simonmar Priority: normal | Status: new Component: Runtime System | Milestone: Keywords: | Version: Architecture: Unknown/Multiple | Operating System: Difficulty: Project (more than a week) | Unknown/Multiple Blocked By: | Type of failure: Related Tickets: 635, 7353 | None/Unknown | Test Case: | Blocking: ----------------------------------------------+---------------------------- This is mainly a reference discussion ticket. libuv (https://github.com/joyent/libuv) is a lightweight library which allows asynchronous IO across OpenBSD, Linux, Darwin, Windows etc by utilizing the fastest implementation on each system (epoll, kqueue, IOCP, event ports). These specialized IO polling methods are '''much''' faster on their respective platforms than just using select() like the RTS currently does. Additionally, it also provides cross platform threads, mutex, condition vars, terminal input/output and term settings w/ cross platform ANSI escape code handling, thread pools, cross platform HRC's etc. It's currently significantly faster than libevent and slightly faster than libev (and no doubt faster than rolling our own stuff). Because it's maintained and utilized heavily by Node.js it's in extremely active and maintained development. Rewriting a portion of the RTS to utilize libuv would have the following benefits; * We could ditch basically all of our platform specific code. Everything under rts/win32 and rts/posix, except the SEH stuff, could be deleted. * libuv is tuned for speed on each platform. This would be an optimization to all our async IO stuff. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8400> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 29

[GHC] #8578: Improvements to SpinLock implementation
by GHC 24 Sep '18

24 Sep '18

#8578: Improvements to SpinLock implementation ------------------------------------+------------------------------------- Reporter: parcs | Owner: parcs Type: task | Status: new Priority: normal | Milestone: Component: Runtime System | 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: | ------------------------------------+------------------------------------- The SpinLock implementation has a number of deficiencies. Here is a pair of patches that improves its implementation. Let me know what you think. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8578> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 11

[GHC] #8281: The impossible happened: primRepToFFIType
by GHC 21 Sep '18

21 Sep '18

#8281: The impossible happened: primRepToFFIType -----------------------------------+--------------------------------------- Reporter: tibbe | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.6.2 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Compile-time crash Unknown/Multiple | Test Case: Difficulty: Unknown | Blocking: Blocked By: | Related Tickets: | -----------------------------------+--------------------------------------- I ran into this error while trying to use GHCi on the hashable package: {{{ $ cabal repl Preprocessing library hashable-1.2.0.10... GHCi, version 7.6.2: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... done. Loading package array-0.4.0.1 ... linking ... done. Loading package deepseq-1.3.0.1 ... linking ... done. Loading package bytestring-0.10.0.2 ... linking ... done. Loading package text-0.11.3.1 ... linking ... done. Loading object (static) dist/build/cbits/fnv.o ... done Loading object (static) dist/build/cbits/getRandomBytes.o ... done final link ... done [1 of 4] Compiling Data.Hashable.RandomSource ( Data/Hashable/RandomSource.hs, interpreted ) [flags changed] [2 of 4] Compiling Data.Hashable.Class ( Data/Hashable/Class.hs, interpreted ) [flags changed] ghc: panic! (the 'impossible' happened) (GHC version 7.6.2 for x86_64-apple-darwin): primRepToFFIType Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8281> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 37