June 2018 - ghc-tickets

Re: [GHC] #7411: Exceptions are optimized away in certain situations
by GHC 28 Jun '18

28 Jun '18

#7411: Exceptions are optimized away in certain situations -------------------------------------+------------------------------------- Reporter: SimonHengel | Owner: tdammers Type: bug | Status: new Priority: high | Milestone: 8.8.1 Component: Compiler | Version: 7.6.1 Resolution: | Keywords: seq, deepseq, | evaluate, exceptions Operating System: Linux | Architecture: x86_64 | (amd64) Type of failure: Incorrect result | Test Case: at runtime | simplCore/should_fail/T7411 Blocked By: | Blocking: Related Tickets: #5129 #15225 | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by tdammers): Replying to [comment:44 simonpj]: > * As I wrote above, the motivation for the state hack is functions lie > {{{ > f :: [Int] -> IO () > f xs = print ys >> print xs > where > ys = reverse xs > > test 0 = return () > test n = f [1,n] >> test (n-1) > }}} > which I expected to be much less efficient without the state hack. Would it be worth trying to demonstrate a program that does get worse in this way? I'm still amazed at how good the results are! OK, I did some experimenting, and it seems that the state hack makes absolutely no difference for such programs. Specifically, I used the following fully-fledged example program: {{{#!hs module Main where import System.Environment import System.IO f :: Handle -> [Int] -> IO () f h xs = hPrint h ys >> hPrint h xs where ys = reverse xs test :: Int -> Handle -> IO () test 0 h = return () test n h = f h [1,n] >> test (n-1) h main = do args <- getArgs case args of x:y:_ -> withFile y WriteMode $ test (read x) x:_ -> test (read x) stdout _ -> test 0 stdout }}} This allows us to pass an arbitrary number of rounds to test on the command line, as well as an output file; passing `/dev/null` makes it easier to run the whole thing through `time`. Now given two GHC builds in `ghc-prof` (normal profiling build) and `ghc-prof-nsh` (GHC and boot libs built with `-fno-state-hack`), and with our example program in `bad- ex.hs`, I can run the following session: {{{ tobias@zoidberg:~/well-typed/devel/ghc-7411/ > ./ghc-prof/inplace/bin/ghc- stage2 bad-ex.hs -fforce-recomp -O2 && time ./bad-ex 50000000 /dev/null [1 of 1] Compiling Main ( bad-ex.hs, bad-ex.o ) Linking bad-ex ... ./bad-ex 50000000 /dev/null 53.50s user 0.20s system 99% cpu 53.713 total tobias@zoidberg:~/well-typed/devel/ghc-7411/ > ./ghc-prof-nsh/inplace/bin /ghc-stage2 bad-ex.hs -fforce-recomp -fno-state-hack -O2 && time ./bad-ex 50000000 /dev/null [1 of 1] Compiling Main ( bad-ex.hs, bad-ex.o ) Linking bad-ex ... ./bad-ex 50000000 /dev/null 52.75s user 0.19s system 99% cpu 52.956 total }}} Which suggests that even for a program that is presumably about as bad as it gets, the state hack doesn't make a difference for the better. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7411#comment:46> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #10789: Notify user when a kind mismatch holds up a type family reduction
by GHC 28 Jun '18

28 Jun '18

#10789: Notify user when a kind mismatch holds up a type family reduction -------------------------------------+------------------------------------- Reporter: goldfire | Owner: Type: feature | Status: new request | 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: Differential Revisions: | -------------------------------------+------------------------------------- Consider this contrived example: {{{ {-# LANGUAGE TypeFamilies, PolyKinds, UndecidableInstances #-} module Bug where import Data.Proxy type family F (a :: k) :: k type instance F a = G a type family G a type instance G a = a foo :: Proxy (F Maybe) -> Proxy Maybe foo = id }}} This (correctly) fails to compile. The error message is {{{ Bug.hs:14:7: Couldn't match type ‘F Maybe’ with ‘Maybe’ Expected type: Proxy (F Maybe) -> Proxy Maybe Actual type: Proxy Maybe -> Proxy Maybe In the expression: id In an equation for ‘foo’: foo = id Failed, modules loaded: none. }}} But this is peculiar, but it surely looks like `F` should be a type-level identity function! Of course, upon further inspection, we see that `F` is partial. It reduces only at kind `*`. This is quite hard to figure out, though, especially given that we're using the "default to `*`" behavior of open type families to arrange for this kind restriction. Thus, I propose: figure out when a type family reduction is held up due to a kind mismatch, and inform the user. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10789> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 15

[GHC] #12922: Kind classes compile with PolyKinds
by GHC 27 Jun '18

27 Jun '18

#12922: Kind classes compile with PolyKinds -------------------------------------+------------------------------------- Reporter: Tritlo | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 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: -------------------------------------+------------------------------------- I was asking around on #haskell to better understand the new language extension, -XTypeInType, and how it is different from -XPolyKinds. To study it, I was working with the following small example: {{{ {-# LANGUAGE TypeFamilies, GADTs, TypeOperators, DataKinds, PolyKinds #-} module Main where import Data.Kind (Type) -- Define a Type for the natural numbers, Zero and a successor data Nat = Z | S Nat class Addable k where type (a :: k) + (b :: k) :: k instance Addable Nat where type (Z + a) = a }}} (for more context, see https://gist.github.com/Tritlo/ce5510e80935ac398a33934ee47c7930) Since according to a responder on #haskell, the ability to have kind classes required TypeInType, and should not work with PolyKinds. As the documentation mentions that there could be leakage between PolyKinds and TypeInType and to report such leaks, I felt that I should report it. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12922> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 11

[GHC] #9383: GHCi parser doesn't properly handle -fobject-code
by GHC 27 Jun '18

27 Jun '18

#9383: GHCi parser doesn't properly handle -fobject-code -------------------------------------+------------------------------------- Reporter: crockeea | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: GHCi | Version: 7.8.2 Keywords: | Operating System: Architecture: Unknown/Multiple | Unknown/Multiple Difficulty: Unknown | Type of failure: Other Blocked By: | Test Case: Related Tickets: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- My complete Haskell source: module Foo where f x = x Shell: $ ghci Foo.hs -fobject-code GHCi, version 7.8.2: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... done. target ‘-fobject-code’ is not a module name or a source file Ok, modules loaded: Linker. Prelude Linker> Despite the wanring/error, GHCi *does* generate the object code. This error also prints when I *load* the compiled Foo object file. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9383> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

Re: [GHC] #4806: Make error message more user friendly when module is not found because package is unusable
by GHC 27 Jun '18

27 Jun '18

#4806: Make error message more user friendly when module is not found because package is unusable -------------------------------------+------------------------------------- Reporter: mitar | Owner: (none) Type: feature request | Status: closed Priority: normal | Milestone: 8.6.1 Component: Compiler | Version: 7.0.1 Resolution: fixed | Keywords: newcomer Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Phab:D4783 Wiki Page: | -------------------------------------+------------------------------------- Changes (by bgamari): * status: merge => closed * resolution: => fixed * milestone: => 8.6.1 Comment: This is already present in 8.6. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/4806#comment:37> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #15317: GHCi panic when trying to avoid GHC_OPTIONS -O warning
by GHC 27 Jun '18

27 Jun '18

#15317: GHCi panic when trying to avoid GHC_OPTIONS -O warning -------------------------------------+------------------------------------- Reporter: ChaiTRex | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: 7.10.4 Component: GHCi | Version: 7.10.3 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: -------------------------------------+------------------------------------- Following the advice on [https://stackoverflow.com/a/27881726/7208029 an answer to "How can I load optimized code in GHCI?"], I get a GHCi panic. Running `ghc Luhn` succeeds just fine. On running `ghci Luhn`, I get: {{{ $ ghci Luhn GHCi, version 7.10.3: http://www.haskell.org/ghc/ :? for help [1 of 1] Compiling Luhn ( Luhn.hs, interpreted ) ghc: panic! (the 'impossible' happened) (GHC version 7.10.3 for x86_64-unknown-linux): floatExpr tick break<15>() Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug }}} Contents of `Luhn.hs`: {{{#!hs {-# OPTIONS_GHC -fobject-code -O3 #-} module Luhn (checkLuhn) where import Data.Bits (shiftL) import Data.Char (digitToInt) import Data.List (foldl') -- Quickly gets a list of digits from a nonnegative Integer -- Gives error for negative inputs -- Uses GMP's show for greatly-improved speed over GMP's div and mod toDigits :: Integer -> [Int] {-# INLINE toDigits #-} toDigits = map digitToInt . show -- Quickly gets the same result as iteratively getting the digit sum of a nonnegative Int until the digit sum is only one digit long -- Gives an erroneous value for negative inputs repeatedDigitSum :: Int -> Int {-# INLINE repeatedDigitSum #-} repeatedDigitSum n = (n - 1) `rem` 9 + 1 -- Gets the Luhn sum, which is zero for valid inputs, of a list of digits -- Uses Data.Bits.shiftL to quickly double luhnSum :: [Int] -> Int {-# INLINE luhnSum #-} luhnSum = fromInteger . flip rem 10 . foldl' (+) 0 . zipWith ($) (cycle [toInteger, toInteger . repeatedDigitSum . flip shiftL 1]) -- Checks whether a nonnegative Integer passes the Luhn algorithm -- Negative inputs are False, since the Luhn algorithm is intended for unsigned inputs checkLuhn :: Integer -> Bool {-# INLINABLE checkLuhn #-} checkLuhn n = (n >= 0) && ((== 0) . luhnSum . reverse . toDigits) n }}} Strangely, `ghci -fobject-code -O3 Luhn` works just great, so apparently it's not a problem with the switches? -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/15317> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #10965: GHC Panic on import with 'OPTIONS_GHC -fobject-code -O'
by GHC 27 Jun '18

27 Jun '18

#10965: GHC Panic on import with 'OPTIONS_GHC -fobject-code -O' --------------------------------------+--------------------------------- Reporter: Orome | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.2 Keywords: | Operating System: MacOS X Architecture: x86_64 (amd64) | Type of failure: None/Unknown Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: --------------------------------------+--------------------------------- I get {{{ > :l Test.hs [1 of 1] Compiling Main ( Test.hs, interpreted ) ghc: panic! (the 'impossible' happened) (GHC version 7.10.2 for x86_64-apple-darwin): floatExpr tick break<11>() Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug }}} with Test.hs: {{{ {-# OPTIONS_GHC -fobject-code -O #-} import Control.Exception (assert) import Data.Char (ord) f :: String -> String f s = assert (all (`elem` letters) s) $ (letters!!) <$> (ix <$> s) where ix ch = (ord ch - ord 'A') letters = ['A'..'Z'] }}} OSX 10.11, GHC 7.10.2; uad-core 64-bit haswell; Homebrew GHC -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10965> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #15221: failure to build pandoc on Debian armhf
by GHC 27 Jun '18

27 Jun '18

#15221: failure to build pandoc on Debian armhf --------------------------------+--------------------------------- Reporter: clint | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: 8.6.1 Component: Compiler | Version: 8.2.2 Keywords: | Operating System: Linux Architecture: arm | Type of failure: None/Unknown Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: --------------------------------+--------------------------------- {{{ [112 of 131] Compiling Text.Pandoc.Readers.HTML ( src/Text/Pandoc/Readers/HTML.hs, dist- ghc/build/Text/Pandoc/Readers/HTML.p_o ) ghc: panic! (the 'impossible' happened) (GHC version 8.2.2 for arm-unknown-linux): piResultTy Addr# String Call stack: CallStack (from HasCallStack): prettyCurrentCallStack, called at compiler/utils/Outputable.hs:1133:58 in ghc:Outputable callStackDoc, called at compiler/utils/Outputable.hs:1137:37 in ghc:Outputable pprPanic, called at compiler/types/Type.hs:949:35 in ghc:Type Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug /usr/share/cdbs/1/class/hlibrary.mk:147: recipe for target 'build-ghc- stamp' failed }}} https://buildd.debian.org/status/fetch.php?pkg=pandoc&arch=armhf&ver=2.2.1-… -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/15221> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

Re: [GHC] #4806: Make error message more user friendly when module is not found because package is unusable
by GHC 26 Jun '18

26 Jun '18

#4806: Make error message more user friendly when module is not found because package is unusable -------------------------------------+------------------------------------- Reporter: mitar | Owner: (none) Type: feature request | Status: merge Priority: normal | Milestone: Component: Compiler | Version: 7.0.1 Resolution: | Keywords: newcomer Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Phab:D4783 Wiki Page: | -------------------------------------+------------------------------------- Changes (by sgillespie): * status: new => merge -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/4806#comment:36> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #13790: GHC doesn't reduce type family in kind signature unless its arm is twisted
by GHC 26 Jun '18

26 Jun '18

#13790: GHC doesn't reduce type family in kind signature unless its arm is twisted -------------------------------------+------------------------------------- Reporter: RyanGlScott | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.2.1-rc2 (Type checker) | Keywords: TypeInType, | Operating System: Unknown/Multiple TypeFamilies | Architecture: | Type of failure: GHC rejects Unknown/Multiple | valid program Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Here's some code (inspired by Richard's musings [https://github.com/goldfirere/singletons/issues/150#issuecomment-306088297 here]) which typechecks with GHC 8.2.1 or HEAD: {{{#!hs {-# LANGUAGE FlexibleInstances #-} {-# LANGUAGE GADTs #-} {-# LANGUAGE TemplateHaskell #-} {-# LANGUAGE TypeFamilies #-} {-# LANGUAGE TypeInType #-} module Bug where import Data.Kind data family Sing (a :: k) data SomeSing (k :: Type) where SomeSing :: Sing (a :: k) -> SomeSing k type family Promote k :: Type class (Promote (Demote k) ~ k) => SingKind (k :: Type) where type Demote k :: Type fromSing :: Sing (a :: k) -> Demote k toSing :: Demote k -> SomeSing k type family DemoteX (a :: k) :: Demote k type instance DemoteX (a :: Type) = Demote a type instance Promote Type = Type instance SingKind Type where type Demote Type = Type fromSing = error "fromSing Type" toSing = error "toSing Type" ----- data N = Z | S N data instance Sing (z :: N) where SZ :: Sing Z SS :: Sing n -> Sing (S n) type instance Promote N = N instance SingKind N where type Demote N = N fromSing SZ = Z fromSing (SS n) = S (fromSing n) toSing Z = SomeSing SZ toSing (S n) = case toSing n of SomeSing sn -> SomeSing (SS sn) }}} Things get more interesting if you try to add this type instance at the end of this file: {{{#!hs type instance DemoteX (n :: N) = n }}} Now GHC will complain: {{{ GHCi, version 8.2.0.20170522: http://www.haskell.org/ghc/ :? for help Loaded GHCi configuration from /home/rgscott/.ghci [1 of 1] Compiling Bug ( Bug.hs, interpreted ) Bug.hs:49:34: error: • Expected kind ‘Demote N’, but ‘n’ has kind ‘N’ • In the type ‘n’ In the type instance declaration for ‘DemoteX’ | 49 | type instance DemoteX (n :: N) = n | ^ }}} That error message smells funny, since we do have a type family instance that says `Demote N = N`! In fact, if you use Template Haskell to split up the declarations manually: {{{#!hs ... instance SingKind N where type Demote N = N fromSing SZ = Z fromSing (SS n) = S (fromSing n) toSing Z = SomeSing SZ toSing (S n) = case toSing n of SomeSing sn -> SomeSing (SS sn) $(return []) type instance DemoteX (n :: N) = n }}} Then the file typechecks without issue. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13790> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3