May 2018 - ghc-tickets

[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] #14366: Type family equation refuses to unify wildcard type patterns
by GHC 05 Jan '19

05 Jan '19

#14366: Type family equation refuses to unify wildcard type patterns -------------------------------------+------------------------------------- Reporter: RyanGlScott | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.2.1 (Type checker) | Keywords: TypeFamilies, | Operating System: Unknown/Multiple TypeInType | Architecture: | Type of failure: GHC rejects Unknown/Multiple | valid program Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- This typechecks: {{{#!hs {-# LANGUAGE TypeFamilies #-} {-# LANGUAGE TypeInType #-} {-# LANGUAGE TypeOperators #-} import Data.Kind import Data.Type.Equality type family Cast (e :: a :~: b) (x :: a) :: b where Cast Refl x = x }}} However, if you try to make the kinds `a` and `b` explicit arguments to `Cast`: {{{#!hs {-# LANGUAGE TypeFamilies #-} {-# LANGUAGE TypeInType #-} {-# LANGUAGE TypeOperators #-} import Data.Kind import Data.Type.Equality type family Cast (a :: Type) (b :: Type) (e :: a :~: b) (x :: a) :: b where Cast _ _ Refl x = x }}} Then GHC gets confused: {{{ GHCi, version 8.2.1: http://www.haskell.org/ghc/ :? for help Loaded GHCi configuration from /home/ryanglscott/.ghci [1 of 1] Compiling Main ( Bug.hs, interpreted ) Bug.hs:9:12: error: • Expected kind ‘_ :~: _1’, but ‘'Refl’ has kind ‘_ :~: _’ • In the third argument of ‘Cast’, namely ‘Refl’ In the type family declaration for ‘Cast’ | 9 | Cast _ _ Refl x = x | ^^^^ }}} A workaround is to explicitly write out what should be inferred by the underscores: {{{#!hs {-# LANGUAGE TypeFamilies #-} {-# LANGUAGE TypeInType #-} {-# LANGUAGE TypeOperators #-} import Data.Kind import Data.Type.Equality type family Cast (a :: Type) (b :: Type) (e :: a :~: b) (x :: a) :: b where Cast a a Refl x = x }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14366> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 8

[GHC] #11829: C++ does not catch exceptions when used with Haskell-main and linked by ghc
by GHC 05 Jan '19

05 Jan '19

#11829: C++ does not catch exceptions when used with Haskell-main and linked by ghc -------------------------------------+------------------------------------- Reporter: pl | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Runtime | Version: 7.10.3 System | Keywords: c++ | Operating System: Unknown/Multiple exceptions | Architecture: | Type of failure: Runtime crash Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Consider the C++ program {{{ // Main_cc.cc #include <iostream> #include <exception> extern "C" { int func() { try { throw std::runtime_error("THIS FAILS!"); } catch(const std::runtime_error &c) { std::cerr << c.what(); } return 0; } } }}} and the Haskell program {{{#!hs -- Main.hs module Main where import Foreign.C.Types foreign import ccall func :: IO CInt main :: IO () main = print =<< func }}} When compiled with {{{ ghc --make Main.hs Main_cc.cc -lstdc++ }}} and being run on mac os x 10.11.3 the result is {{{ libc++abi.dylib: terminating with uncaught exception of type std::runtime_error: THIS FAILS! Abort trap: 6 }}} and not (as it should be): {{{ THIS FAILS!0 }}} When compiled in a similar way under windows some c++ exceptions of a larger project are not caught as well. However, these were hard to track down and I was unable to construct a simple example like the one above. As a workaround one can * write a main() function in c++ and call from there into Haskell, * compile the Haskell-Module with the options "-c -no-hs-main", and * link the application using g++ (not ghc!). When being compiled / linked this way, all c++ exceptions are being caught in mac os x as well as under windows. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11829> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #13630: ByteString pinned memory can be leaky
by GHC 04 Jan '19

04 Jan '19

#13630: ByteString pinned memory can be leaky -------------------------------------+------------------------------------- Reporter: nh2 | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Runtime | Version: 8.0.1 System | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- My question on IRC: {{{ How does memory allocation for pinned blocks work? Let's say pinned blocks are 4KB in size, and I allocate first a 3 KB ByteString A and then an 8-byte ByteString B. Now I GC A, no longer need it. Then according to https://ghc.haskell.org/trac/ghc/wiki/Commentary/Rts/Storage/GC/Pinned "a single pinned object keeps alive the whole block in which it resides", my small ByteString B keeps the entire block alive. But what happens with the 3KB in the front of that block? Will it be re-used by the next ByteString allocation (say 1KB)? In other words, how smart is allocatePinned as an allocator? }}} The answer: {{{ slyfox: nh2: allocatePinned it quite dump. it only allocated from free tail space }}} {{{ nh2: slyfox: that sounds like a huge potential for memory leak then slyfox: yes, fragmentation is quite bad for bytestrings }}} So it seems that I can get into the unfortunate situation where a super short `ByteString` of a few bytes can waste an entire 4 KB block of memory; some migth call this a leak. One idea to solve it seems to be to change standard `ByteStrings` to not pinned, and to allocate pinned ones explicitly when needed. This seems to be an often-discussed topic and not trivial because many `ByteString` functions are implemented using libc FFI functions. However, it seems there will always be _some_ need for pinned memory, so we should better have an efficient way to manage it in any case. Efficient here means, for example, to re-use freed memory inside a block instead of only using free tail space. It seems that `jemalloc` has a feature to use given blocks of memory and provide a `malloc()` functionality inside them: https://stackoverflow.com/questions/30836359/jemalloc-mmap-and-shared- memory Perhaps this could be used to provide GHC with a simple method to use pinned memory more efficiently? -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13630> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #14797: High-residency modules during GHC build
by GHC 04 Jan '19

04 Jan '19

#14797: High-residency modules during GHC build -------------------------------------+------------------------------------- Reporter: tdammers | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.2.1 Keywords: | Operating System: Linux Architecture: x86_64 | Type of failure: Runtime (amd64) | performance bug Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- While trying to figure out memory allocation failures on the CI server, I tracked allocations and residency on a per-module basis, and noticed that some modules require a lot more memory to compile than others. I've compiled a digest of the most memory-hungry compilation steps: {{{ "/usr/local/bin/ghc" ... compiler/stage1/build/OptCoercion.o: Max. residency: 991356584 / In use: 2094M "inplace/bin/ghc-stage1" ... compiler/stage2/build/OptCoercion.dyn_o: Max. residency: 714940528 / In use: 1471M "/usr/local/bin/ghc" ... compiler/stage1/build/X86/CodeGen.o: Max. residency: 679947656 / In use: 1497M "inplace/bin/ghc-stage1" ... libraries/Cabal/Cabal/dist- install/build/Distribution/SPDX/LicenseId.dyn_o: Max. residency: 653177496 / In use: 1355M "inplace/bin/ghc-stage1" ... compiler/stage2/build/X86/CodeGen.dyn_o: Max. residency: 641515344 / In use: 1425M "/usr/local/bin/ghc" ... -ilibraries/parsec: Max. residency: 593335568 / In use: 1355M "/usr/local/bin/ghc" ... libraries/Cabal/Cabal/dist- boot/build/Distribution/SPDX/LicenseId.o: Max. residency: 575034968 / In use: 1137M "inplace/bin/ghc-stage1" ... compiler/stage2/build/DynFlags.dyn_o: Max. residency: 402734408 / In use: 1059M "/usr/local/bin/ghc" ... compiler/stage1/build/DynFlags.o: Max. residency: 343154784 / In use: 892M "inplace/bin/ghc-stage1" ... libraries/template-haskell/dist- install/build/Language/Haskell/TH/Syntax.dyn_o: Max. residency: 266744312 / In use: 720M "/usr/local/bin/ghc" ... compiler/stage1/build/Parser.o: Max. residency: 258613584 / In use: 679M "inplace/bin/ghc-stage1" ... compiler/stage2/build/CmmOpt.dyn_o: Max. residency: 254639624 / In use: 568M "inplace/bin/ghc-stage1" ... compiler/stage2/build/Parser.dyn_o: Max. residency: 236774592 / In use: 635M "inplace/bin/ghc-stage1" ... ghc/stage2/build/GHCi/UI.dyn_o: Max. residency: 234055792 / In use: 609M "/usr/local/bin/ghc" ... compiler/stage1/build/HsExpr.o: Max. residency: 233724400 / In use: 657M "/usr/local/bin/ghc" ... compiler/stage1/build/CmmOpt.o: Max. residency: 208378888 / In use: 436M "/usr/local/bin/ghc" ... compiler/stage1/build/HscMain.o: Max. residency: 197177248 / In use: 500M "inplace/bin/ghc-stage1" ... compiler/stage2/build/HsExpr.dyn_o: Max. residency: 192780496 / In use: 521M "inplace/bin/ghc-stage1" ... libraries/Cabal/Cabal/dist- install/build/Distribution/Simple/Setup.dyn_o: Max. residency: 191649952 / In use: 521M "/usr/local/bin/ghc" ... compiler/stage1/build/HsDecls.o: Max. residency: 183441120 / In use: 508M "inplace/bin/ghc-stage1" ... compiler/stage2/build/TcSplice.dyn_o: Max. residency: 176802000 / In use: 445M "inplace/bin/ghc-stage1" ... compiler/stage2/build/TcRnDriver.dyn_o: Max. residency: 172892128 / In use: 458M "inplace/bin/ghc-stage1" ... compiler/stage2/build/HscMain.dyn_o: Max. residency: 172802688 / In use: 470M "inplace/bin/ghc-stage1" ... libraries/ghci/dist- install/build/GHCi/TH/Binary.dyn_o: Max. residency: 169775368 / In use: 455M "inplace/bin/ghc-stage1" ... compiler/stage2/build/GhcMake.dyn_o: Max. residency: 167305408 / In use: 431M "inplace/bin/ghc-stage1" ... compiler/stage2/build/PrimOp.dyn_o: Max. residency: 162532408 / In use: 455M "/usr/local/bin/ghc" ... compiler/stage1/build/HscTypes.o: Max. residency: 160814736 / In use: 421M "inplace/bin/ghc-stage1" ... compiler/stage2/build/HsDecls.dyn_o: Max. residency: 160445552 / In use: 426M "inplace/bin/ghc-stage1" ... libraries/Cabal/Cabal/dist- install/build/Language/Haskell/Extension.dyn_o: Max. residency: 159817712 / In use: 363M "/usr/local/bin/ghc" ... compiler/stage1/build/TcSplice.o: Max. residency: 156738656 / In use: 375M "/usr/local/bin/ghc" ... compiler/stage1/build/TcRnDriver.o: Max. residency: 153910768 / In use: 421M "inplace/bin/ghc-stage1" ... libraries/containers/dist- install/build/Data/Sequence/Internal.dyn_o: Max. residency: 151318224 / In use: 405M "inplace/bin/ghc-stage1" ... compiler/stage2/build/MatchLit.dyn_o: Max. residency: 150681176 / In use: 382M "/usr/local/bin/ghc" ... compiler/stage1/build/ByteCodeGen.o: Max. residency: 144579176 / In use: 389M "/usr/local/bin/ghc" ... compiler/stage1/build/GhcMake.o: Max. residency: 144411312 / In use: 379M "/usr/local/bin/ghc" ... compiler/stage1/build/MatchLit.o: Max. residency: 142596736 / In use: 382M "inplace/bin/ghc-stage1" ... compiler/stage2/build/StgCmmPrim.dyn_o: Max. residency: 141484616 / In use: 340M "inplace/bin/ghc-stage1" ... libraries/ghci/dist- install/build/GHCi/Message.dyn_o: Max. residency: 139224136 / In use: 363M "/usr/local/bin/ghc" ... compiler/stage1/build/StgCmmPrim.o: Max. residency: 137390744 / In use: 346M "inplace/bin/ghc-stage1" ... compiler/stage2/build/TcExpr.dyn_o: Max. residency: 136387128 / In use: 370M "inplace/bin/ghc-stage1" ... compiler/stage2/build/Lexer.dyn_o: Max. residency: 135228120 / In use: 337M "inplace/bin/ghc-stage1" ... compiler/stage2/build/TcTyClsDecls.dyn_o: Max. residency: 134817024 / In use: 369M "/usr/local/bin/ghc" ... compiler/stage1/build/MkIface.o: Max. residency: 134039144 / In use: 339M "inplace/bin/ghc-stage1" ... compiler/stage2/build/RnExpr.dyn_o: Max. residency: 133947600 / In use: 345M "inplace/bin/ghc-stage1" ... compiler/stage2/build/HscTypes.dyn_o: Max. residency: 133938840 / In use: 366M "inplace/bin/ghc-stage1" ... compiler/stage2/build/TcGenDeriv.dyn_o: Max. residency: 133744056 / In use: 320M "/usr/local/bin/ghc" ... compiler/stage1/build/TcExpr.o: Max. residency: 131766896 / In use: 336M "inplace/bin/ghc-stage1" ... compiler/stage2/build/GHC.dyn_o: Max. residency: 130998912 / In use: 277M "inplace/bin/ghc-stage1" ... compiler/stage2/build/ByteCodeGen.dyn_o: Max. residency: 129499464 / In use: 348M "inplace/bin/ghc-stage1" ... compiler/stage2/build/InteractiveEval.dyn_o: Max. residency: 128902944 / In use: 286M }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14797> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #11962: Support induction recursion
by GHC 04 Jan '19

04 Jan '19

#11962: Support induction recursion -------------------------------------+------------------------------------- Reporter: goldfire | Owner: Type: feature | Status: new request | Priority: normal | Milestone: Component: Compiler | Version: 8.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Now that we have `-XTypeInType`, let's take it a step further and allow induction recursion. This feature exists in Agda and Idris. Here is a short example of what can be done in Agda: {{{ mutual -- Codes for types. data U : Set where nat : U pi : (a : U) → (El a → U) → U -- A function that interprets codes as actual types. El : U → Set El nat = ℕ El (pi a b) = (x : El a) → El (b x) }}} Note that the `U` datatype and the `El` function depend on each other. But if you look more closely, the header for `U` does not depend on `El`; only the constructors of `U` depend on `El`. So if we typecheck `U : Set` first, then `El : U → Set`, then the constructors of `U`, then the equations of `El`, we're OK. Translation into Haskell: {{{ import Data.Kind data family Sing (a :: k) -- we still require singletons data U :: Type where Nat :: U Pi :: Sing (a :: U) -> (El a -> U) -> U type family El (u :: U) :: Type where El 'Nat = Int El (Pi a b) = forall (x :: El a). Sing x -> El (b x) }}} This currently errors with {{{ • Type constructor ‘U’ cannot be used here (it is defined and used in the same recursive group) • In the kind ‘U’ }}} It needn't error. (I'm cheating a bit here, because for unrelated reasons, we can't return a `forall` on the right-hand side of a type family. But that's not the issue at hand.) I have some very sketchy notes on how we might do this [wiki:DependentHaskell/Internal#Separatingtypesignaturesfromdefinitions here]. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11962> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 12

[GHC] #15155: How untagged pointers sneak into banged fields
by GHC 03 Jan '19

03 Jan '19

#15155: How untagged pointers sneak into banged fields -------------------------------------+------------------------------------- Reporter: heisenbug | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: 8.6.1 Component: Compiler | Version: 8.4.2 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: 14677 | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- (''N.B.'' I am writing this up from memory, and cannot verify it just now, maybe someone can lend a hand, otherwise I'll do it ASAP!) Here is a way how untagged pointers to strict data can be created in banged (strict) constructor fields. This reproduction recipe **depends on the patch from #14677 applied**. We have 3 modules `A`, `B` and `C`: {{{#!hs module A where data A = X | Y | Z a = Z }}} {{{#!hs module B where import A newtype B = B A b = B a }}} {{{#!hs {-# language MagicHash #-} module C where import A import B import GHC.Exts data C = C !B c = C b main = do print (I# (reallyUnsafePtrEquality# a (coerce b))) -- prints 0, b is softlink print (I# (dataToTag# c)) -- prints 0: not entered yet print (case c of C b' -> I# (dataToTag# b')) -- prints 0? print (case c of C (B a') -> I# (dataToTag# a')) -- prints 3 }}} ------------------- == Why this happens `B.b` is a newtype to `A.a` so one would expect that both alias the same memory location (a ''hardlink'' in filesystem parlance). But currently reexports are implemented with a special type of closure `IND_STATIC` (a ''softlink'') which needs to be entered to obtain the actual (tagged pointer). The `IND_STATIC` closure's pointer is never tagged (otherwise it would never be entered, instead interpreted as a honest-to-goodness `A.A`, which causes the symptoms seen in #14677). With #14677 applied to GHC, the unfolding of `B.b` is correctly read when compiling `C` (with `-O1` and better) and thus the compiler knows that it should be a tagged pointer value. Thus the construction of `C.c` shortcuts the entering of `B.b` when filling the strict field, and (because `B.b` being a softlink, thus untagged) the field ends up carrying a 0 tag. ------------------------- == How can this be fixed? I see two possibilities one conservative and one invasive. === Conservative When seeing a coercion unfolding of a tagged value being used to initialise a strict field, do not skip the evaluatedness check, but cater for the possibility of an `IND_STATIC` closure. Check the closure type, and if confirmed, steal the pointee and use that. === Invasive Get rid of the `IND_STATIC` closures altogether. For ''intra-module'' softlinks we can have proper hardlinks (assembler `.equiv` directives, or LLVM `alias`es). ''Inter-module'' softlinks can also be eliminated by linker scripts. This would however cause more build artifacts, so I don't know how hairy it would turn out. OTOH, it would reduce binary size by eliminating indirection closures and potential dereferencing code. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/15155> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 18

[GHC] #8611: nofib’s cacheprof’s allocations nondeterminisitic
by GHC 01 Jan '19

01 Jan '19

#8611: nofib’s cacheprof’s allocations nondeterminisitic -----------------------------------------+--------------------------------- Reporter: nomeata | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: NoFib benchmark suite | Version: 7.6.3 Keywords: | Operating System: Architecture: Unknown/Multiple | Unknown/Multiple Difficulty: Unknown | Type of failure: None/Unknown Blocked By: | Test Case: Related Tickets: | Blocking: -----------------------------------------+--------------------------------- This seems to be neither expected nor desired, and hence worth investigating: {{{ ./cacheprof +RTS -t <<ghc: 127588752 bytes, 243 GCs, 2261288/6551032 avg/max bytes residency (6 samples), 15M in use, 0.00 INIT (0.00 elapsed), 0.12 MUT (0.12 elapsed), 0.05 GC (0.05 elapsed) :ghc>> ./cacheprof +RTS -t <<ghc: 127392144 bytes, 242 GCs, 2261304/6551576 avg/max bytes residency (6 samples), 15M in use, 0.00 INIT (0.00 elapsed), 0.11 MUT (0.11 elapsed), 0.04 GC (0.05 elapsed) :ghc>> }}} The allocations numbers change, when they shoudn’t. Reproduced with 7.6.3 and current HEAD. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8611> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 16

[GHC] #14974: 2-fold memory usage regression GHC 8.2.2 -> GHC 8.4.1 compiling `mmark` package
by GHC 30 Dec '18

30 Dec '18

#14974: 2-fold memory usage regression GHC 8.2.2 -> GHC 8.4.1 compiling `mmark` package -------------------------------------+------------------------------------- Reporter: hvr | Owner: (none) Type: bug | Status: new Priority: high | Milestone: Component: Compiler | Version: 8.4.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Compile-time Unknown/Multiple | performance bug Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- I haven't had time yet to diagnose where the memory is going to, but here's how to reproduce: {{{ cabal get mmark-0.0.5.6 && cd mmark-0.0.5.6/ cat > cabal.project <<EOF packages: . package mmark ghc-options: -Rghc-timing EOF }}} Then, `cabal new-build -w ghc-8.4.1` and ``cabal new-build -w ghc-8.2.2` will respectively output {{{ Resolving dependencies... Build profile: -w ghc-8.4.1 -O1 In order, the following will be built (use -v for more details): - mmark-0.0.5.6 (lib) (first run) Configuring library for mmark-0.0.5.6.. <<ghc: 13510872 bytes, 13 GCs, 806197/1299656 avg/max bytes residency (3 samples), 4M in use, 0.001 INIT (0.000 elapsed), 0.007 MUT (0.025 elapsed), 0.018 GC (0.018 elapsed) :ghc>> Preprocessing library for mmark-0.0.5.6.. Building library for mmark-0.0.5.6.. [1 of 9] Compiling Text.MMark.Parser.Internal.Type ( Text/MMark/Parser/Internal/Type.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.4.1/mmark-0.0.5.6/build/Text/MMark/Parser/Internal/Type.o ) [2 of 9] Compiling Text.MMark.Parser.Internal ( Text/MMark/Parser/Internal.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.4.1/mmark-0.0.5.6/build/Text/MMark/Parser/Internal.o ) [3 of 9] Compiling Text.MMark.Type ( Text/MMark/Type.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.4.1/mmark-0.0.5.6/build/Text/MMark/Type.o ) Text/MMark/Type.hs:38:1: warning: [-Wunused-imports] The import of ‘Data.Semigroup’ is redundant except perhaps to import instances from ‘Data.Semigroup’ To import instances alone, use: import Data.Semigroup() | 38 | import Data.Semigroup | ^^^^^^^^^^^^^^^^^^^^^ [4 of 9] Compiling Text.MMark.Trans ( Text/MMark/Trans.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.4.1/mmark-0.0.5.6/build/Text/MMark/Trans.o ) [5 of 9] Compiling Text.MMark.Util ( Text/MMark/Util.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.4.1/mmark-0.0.5.6/build/Text/MMark/Util.o ) [6 of 9] Compiling Text.MMark.Render ( Text/MMark/Render.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.4.1/mmark-0.0.5.6/build/Text/MMark/Render.o ) Text/MMark/Render.hs:25:1: warning: [-Wunused-imports] The import of ‘Data.Semigroup’ is redundant except perhaps to import instances from ‘Data.Semigroup’ To import instances alone, use: import Data.Semigroup() | 25 | import Data.Semigroup | ^^^^^^^^^^^^^^^^^^^^^ [7 of 9] Compiling Text.MMark.Parser ( Text/MMark/Parser.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.4.1/mmark-0.0.5.6/build/Text/MMark/Parser.o ) [8 of 9] Compiling Text.MMark ( Text/MMark.hs, /tmp/mmark-0.0.5.6 /dist- newstyle/build/x86_64-linux/ghc-8.4.1/mmark-0.0.5.6/build/Text/MMark.o ) [9 of 9] Compiling Text.MMark.Extension ( Text/MMark/Extension.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.4.1/mmark-0.0.5.6/build/Text/MMark/Extension.o ) <<ghc: 89663269464 bytes, 1856 GCs, 338140365/1430147624 avg/max bytes residency (23 samples), 4130M in use, 0.000 INIT (0.000 elapsed), 58.546 MUT (62.124 elapsed), 31.536 GC (31.505 elapsed) :ghc>> <<ghc: 65387008 bytes, 50 GCs, 2189716/5047200 avg/max bytes residency (5 samples), 11M in use, 0.000 INIT (0.000 elapsed), 0.021 MUT (0.207 elapsed), 0.042 GC (0.042 elapsed) :ghc>> }}} and {{{ Resolving dependencies... Build profile: -w ghc-8.4.1 -O1 In order, the following will be built (use -v for more details): - mmark-0.0.5.6 (lib) (first run) Configuring library for mmark-0.0.5.6.. <<ghc: 13510872 bytes, 13 GCs, 806197/1299656 avg/max bytes residency (3 samples), 4M in use, 0.001 INIT (0.000 elapsed), 0.007 MUT (0.025 elapsed), 0.018 GC (0.018 elapsed) :ghc>> Preprocessing library for mmark-0.0.5.6.. Building library for mmark-0.0.5.6.. [1 of 9] Compiling Text.MMark.Parser.Internal.Type ( Text/MMark/Parser/Internal/Type.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.4.1/mmark-0.0.5.6/build/Text/MMark/Parser/Internal/Type.o ) [2 of 9] Compiling Text.MMark.Parser.Internal ( Text/MMark/Parser/Internal.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.4.1/mmark-0.0.5.6/build/Text/MMark/Parser/Internal.o ) [3 of 9] Compiling Text.MMark.Type ( Text/MMark/Type.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.4.1/mmark-0.0.5.6/build/Text/MMark/Type.o ) [4 of 9] Compiling Text.MMark.Trans ( Text/MMark/Trans.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.4.1/mmark-0.0.5.6/build/Text/MMark/Trans.o ) [5 of 9] Compiling Text.MMark.Util ( Text/MMark/Util.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.4.1/mmark-0.0.5.6/build/Text/MMark/Util.o ) [6 of 9] Compiling Text.MMark.Render ( Text/MMark/Render.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.4.1/mmark-0.0.5.6/build/Text/MMark/Render.o ) [7 of 9] Compiling Text.MMark.Parser ( Text/MMark/Parser.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.4.1/mmark-0.0.5.6/build/Text/MMark/Parser.o ) [8 of 9] Compiling Text.MMark ( Text/MMark.hs, /tmp/mmark-0.0.5.6 /dist- newstyle/build/x86_64-linux/ghc-8.4.1/mmark-0.0.5.6/build/Text/MMark.o ) [9 of 9] Compiling Text.MMark.Extension ( Text/MMark/Extension.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.4.1/mmark-0.0.5.6/build/Text/MMark/Extension.o ) <<ghc: 89663269464 bytes, 1856 GCs, 338140365/1430147624 avg/max bytes residency (23 samples), 4130M in use, 0.000 INIT (0.000 elapsed), 58.546 MUT (62.124 elapsed), 31.536 GC (31.505 elapsed) :ghc>> <<ghc: 65387008 bytes, 50 GCs, 2189716/5047200 avg/max bytes residency (5 samples), 11M in use, 0.000 INIT (0.000 elapsed), 0.021 MUT (0.207 elapsed), 0.042 GC (0.042 elapsed) :ghc>> hvr@duo:/tmp/mmark-0.0.5.6$ cabal new-build -w ghc-8.2.2 Resolving dependencies... Build profile: -w ghc-8.2.2 -O1 In order, the following will be built (use -v for more details): - mmark-0.0.5.6 (lib) (first run) Configuring library for mmark-0.0.5.6.. <<ghc: 11200976 bytes, 11 GCs, 853349/1518976 avg/max bytes residency (3 samples), 5M in use, 0.001 INIT (0.000 elapsed), 0.017 MUT (0.037 elapsed), 0.021 GC (0.021 elapsed) :ghc>> Preprocessing library for mmark-0.0.5.6.. Building library for mmark-0.0.5.6.. [1 of 9] Compiling Text.MMark.Parser.Internal.Type ( Text/MMark/Parser/Internal/Type.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.2.2/mmark-0.0.5.6/build/Text/MMark/Parser/Internal/Type.o ) [2 of 9] Compiling Text.MMark.Parser.Internal ( Text/MMark/Parser/Internal.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.2.2/mmark-0.0.5.6/build/Text/MMark/Parser/Internal.o ) [3 of 9] Compiling Text.MMark.Type ( Text/MMark/Type.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.2.2/mmark-0.0.5.6/build/Text/MMark/Type.o ) [4 of 9] Compiling Text.MMark.Trans ( Text/MMark/Trans.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.2.2/mmark-0.0.5.6/build/Text/MMark/Trans.o ) [5 of 9] Compiling Text.MMark.Util ( Text/MMark/Util.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.2.2/mmark-0.0.5.6/build/Text/MMark/Util.o ) [6 of 9] Compiling Text.MMark.Render ( Text/MMark/Render.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.2.2/mmark-0.0.5.6/build/Text/MMark/Render.o ) [7 of 9] Compiling Text.MMark.Parser ( Text/MMark/Parser.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.2.2/mmark-0.0.5.6/build/Text/MMark/Parser.o ) [8 of 9] Compiling Text.MMark ( Text/MMark.hs, /tmp/mmark-0.0.5.6 /dist- newstyle/build/x86_64-linux/ghc-8.2.2/mmark-0.0.5.6/build/Text/MMark.o ) [9 of 9] Compiling Text.MMark.Extension ( Text/MMark/Extension.hs, /tmp/mmark-0.0.5.6/dist- newstyle/build/x86_64-linux/ghc-8.2.2/mmark-0.0.5.6/build/Text/MMark/Extension.o ) <<ghc: 89565902800 bytes, 2004 GCs, 225836231/772393000 avg/max bytes residency (27 samples), 2179M in use, 0.001 INIT (0.000 elapsed), 58.237 MUT (61.472 elapsed), 25.823 GC (25.795 elapsed) :ghc>> <<ghc: 79930048 bytes, 56 GCs, 3522920/10072040 avg/max bytes residency (6 samples), 20M in use, 0.001 INIT (0.000 elapsed), 0.028 MUT (0.213 elapsed), 0.059 GC (0.059 elapsed) :ghc>> }}} Consequently, - GHC 8.2.2: **2179M in use**, 0.001 INIT (0.000 elapsed), 58.237 MUT (61.472 elapsed), 25.823 GC (25.795 elapsed) - GHC 8.4.1: **4130M in use**, 0.000 INIT (0.000 elapsed), 58.546 MUT (62.124 elapsed), 31.536 GC (31.505 elapsed) -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14974> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 14

[GHC] #14672: Make likelyhood of branches/conditions available throughout the compiler.
by GHC 28 Dec '18

28 Dec '18

#14672: Make likelyhood of branches/conditions available throughout the compiler. -------------------------------------+------------------------------------- Reporter: AndreasK | Owner: (none) Type: task | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.2.2 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- For simple code such as `if a > 10 then foo else bar` it can make a big difference if we jump on `a > 10` or `a <= 10`. We already tracked the likelyhood of such conditionals for stack/heap checks in Cmm. I recently updated the native codegen on X86 to take this information into account where available and in one edge case this improved execution time from 1.7s to 1.5s! Having this info available for other cases like bound checking (assume valid) or partial pattern matches (assume no match failure) should be a small win for most programs and hopefully a big one for a few of them. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14672> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 22