- ghc-tickets - Haskell.org

[GHC] #10825: Poor performance of optimized code.
by GHC 17 Sep '15

17 Sep '15

#10825: Poor performance of optimized code. -------------------------------------+------------------------------------- Reporter: AndriusS | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.4 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Runtime Unknown/Multiple | performance bug Test Case: | Blocked By: Blocking: | Related Tickets: #8814, #8835 Differential Revisions: | -------------------------------------+------------------------------------- This is possible duplicate of #8814, #8835. I am creating new ticket as I have fairly small example with no dependencies. The problem is that optimized code runs orders of magnitude slower than non-optimized. {{{#!hs import System.IO.Unsafe (unsafePerformIO) newtype MyList = MyList Int equalsLast :: MyList -> Int -> Bool equalsLast (MyList value) = (== value) fromList :: [Int] -> MyList fromList = MyList . last {- minimizing imports -} (<$>) = fmap replicateM' n x = sequence (replicate n x) readInt = read <$> getLine :: IO Int readAndMap :: Int -> (Int->Bool) -> IO [Bool] readAndMap n f = replicateM' n (f <$> readInt) {- if readAndMap is replaced by commented out line below, the problem goes away -} --readAndMap n f = map f <$> replicateM' n (readInt) fromList' :: [Int] -> MyList {- an attempt to further demostrate the issue. fromList'=fromList is very slow as well -} fromList' list = unsafePerformIO $ do putStrLn "fromList" return $ fromList list readMyList :: Int -> IO MyList readMyList n = fromList' <$> (replicateM' n $ read <$> getLine) main :: IO () main = do n1 <- read <$> getLine :: IO Int n2 <- read <$> getLine :: IO Int l <- readMyList n1 ans <- readAndMap n2 (equalsLast l) mapM_ print ans }}} If this code is compiled with no flags, it prints "fromList" once and runs in 0.3s. Compiled with "-O" prints "fromList" many times and takes over 6 seconds. Here's a Haskell script to generate input for program above: {{{#!hs main = sequence_ $ map print (n:n:[1..n]++[1..n]) where n = 10000 }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10825> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

Re: [GHC] #1168: Optimisation sometimes decreases sharing in IO code
by GHC 17 Sep '15

17 Sep '15

#1168: Optimisation sometimes decreases sharing in IO code -------------------------------------+------------------------------------- Reporter: simonmar | Owner: Type: bug | Status: new Priority: normal | Milestone: ⊥ Component: Compiler | Version: 6.6 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: | nofib/spectral/calendar Blocked By: | Blocking: Related Tickets: | Differential Revisions: -------------------------------------+------------------------------------- Comment (by thomie): See #10825 for another example. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/1168#comment:19> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #7830: Error: operand out of range
by GHC 17 Sep '15

17 Sep '15

#7830: Error: operand out of range -------------------------------------+------------------------------------- Reporter: erikd | Owner: Type: bug | Status: new Priority: highest | Milestone: 8.0.1 Component: Compiler | Version: 7.8.1-rc1 Resolution: | Keywords: Operating System: Linux | Architecture: powerpc Type of failure: Installing GHC | Test Case: failed | Blocked By: | Blocking: Related Tickets: | Differential Revisions: -------------------------------------+------------------------------------- Comment (by erikd): Replying to [comment:34 trommler]: > I could prepare a patch next week and verify on powerpc64. @erikd could you help out building the patch on your machine? Certainly! -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7830#comment:35> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #10885: Crashes on FFI calls to Visual Studio-built C library
by GHC 17 Sep '15

17 Sep '15

#10885: Crashes on FFI calls to Visual Studio-built C library -------------------------------------+------------------------------------- Reporter: mwu | Owner: Type: bug | Status: new Priority: high | Milestone: Component: Compiler | Version: 7.10.1 Keywords: | Operating System: Windows Architecture: x86_64 | Type of failure: Runtime crash (amd64) | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- I want to call third-party C library using ffi on Windows. Everything works fine with 32-bit GHC distribution, however 64-bit program crashes as soon the ffi call is being made. Interestingly, the same program works when run with {{{runghc}}}. Just the compiled .exe crashes. Consider the following C++ library code: {{{#!c++ // adder.cpp extern "C" { __declspec(dllexport) int _cdecl add(int a, int b) { return a + b; } } }}} and the following Haskell code using the library: {{{#!hs import Foreign.C foreign import ccall unsafe "add" add :: CInt -> CInt -> IO CInt main = (add 5 2) >>= print }}} It does work on 32-bit architecture and in ghci (on all architectures). Why does it not work as 64-bit executable? Notes: * it works if the library is built with MinGW. However, this is not an acceptable workaround for me, since the library I want to call is external and closed-source * the problem seems to be related to the import library file handling — if I replace the MSVC-generated one with the MinGW-generated .lib import library, it works (even with MSVC-generated .dll) = Repro = I attach archive {{{repro_src.7z}}} with source files (as presented here) and a {{{build.bat}}} script building binaries. It needs to have both GCC/GHC and MSVC compilers in the path. (so it should be called from MSVC's {{{x64 Native Tools Command Prompt}}}) It creates two folder — in one it builds the library with MinGW, in the another one — with MSVC. The Haskell program crashes only with MSVC build of library. However, the Haskell source file works fine with runghc/ghci. Run the {{{build.bat}}} and ender the bin-msvc directory. The {{{Caller.exe}}} program crashes. However, {{{runghc -ladder ..\src\Caller.hs}}} gives the expected result. If you don't have MSVC installed, I provide also a package with all the sources built — {{{repro_bin.7z}}}. I have used Microsoft Visual Studio 2013 and MinGHC 7.10.1, both 64-bit targeting. Same happens with GHC 7.8.4. Please let me know, if you need any further information to investigate this report. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10885> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 12

[GHC] #10889: Compiled software can not read UTF-8 characters using readFile when run via Upstart
by GHC 17 Sep '15

17 Sep '15

#10889: Compiled software can not read UTF-8 characters using readFile when run via Upstart -------------------------------------+------------------------------------- Reporter: SiXoS | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.2 Keywords: | Operating System: Linux Architecture: | Type of failure: Runtime crash Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- Honestly I dont know if this is a problem with Upstart or ghc. I'm Using Ubuntu 14 Create a software that reads from a file containing UTF-8 characters. Compile the program. Running the program from the terminal works but not if you run it via Upstart. It will crash with the following error: "hGetContents: invalid argument (invalid byte sequence)" Minimal Example: Main.hs: {{{#!hs module Main where main :: IO () main = readFile "ex.txt" >>= putStr }}} ex.txt: {{{ å }}} Place ex.txt and the compiled program "Comp" in the same directory "/the/path/" /etc/init/bug-test.conf: {{{ start on runlevel [2345] stop on runlevel [^2345] chdir /the/path/ exec ./Comp }}} If you run ./Comp from the terminal it works. But if you run "sudo service bug-test start" and then check "/var/log/upstart/bug-test.log" you will see that it has crashed. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10889> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

[GHC] #9710: Clean up Trac versions
by GHC 16 Sep '15

16 Sep '15

#9710: Clean up Trac versions -------------------------------------+------------------------------------- Reporter: goldfire | Owner: Type: task | Status: new Priority: normal | Milestone: Component: Build System | 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: -------------------------------------+------------------------------------- The "Version" pull-down menu in Trac has no apparent ordering principle. This should be fixed. Even better, scrap many of the very old versions, as they're unsupported, anyway. Thanks! -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9710> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

Re: [GHC] #4438: Rename and register the "PArr" language extension when it is judged to be ready
by GHC 16 Sep '15

16 Sep '15

#4438: Rename and register the "PArr" language extension when it is judged to be ready -------------------------------------+------------------------------------- Reporter: duncan | Owner: benl Type: task | Status: closed Priority: normal | Milestone: ⊥ Component: Data Parallel | Version: Haskell | Resolution: fixed | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Revisions: -------------------------------------+------------------------------------- Changes (by thomie): * status: new => closed * resolution: => fixed * blockedby: 7330 => Comment: {{{ $ git grep ParallelArrays Cabal/Language/Haskell/Extension.hs: | ParallelArrays }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/4438#comment:4> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] Batch modify: #1544, #5945, #6092, #7114, #7596, #9157, #9655, ...
by GHC 16 Sep '15

16 Sep '15

Batch modification to #1544, #5945, #6092, #7114, #7596, #9157, #9655, #10371, #10421, #10434, #5262 by thomie: failure to Runtime performance bug -- Tickets URL: <http://ghc.haskell.org/trac/ghc/query?id=1544%2C5945%2C6092%2C7114%2C7596%2…> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #7428: GHC compile times are seriously non-linear in program size
by GHC 16 Sep '15

16 Sep '15

#7428: GHC compile times are seriously non-linear in program size -------------------------------------+------------------------------------- Reporter: nudded | Owner: Type: bug | Status: new Priority: normal | Milestone: 8.0.1 Component: Compiler | Version: 7.4.2 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: Type of failure: Compile-time | Unknown/Multiple performance bug | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Revisions: -------------------------------------+------------------------------------- Changes (by thomie): * failure: Compile-time crash => Compile-time performance bug -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7428#comment:23> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] Batch modify: #6132, #7650, #4176, #8556
by GHC 16 Sep '15

16 Sep '15

Batch modification to #6132, #7650, #4176, #8556 by thomie: component to Compiler (Parser) -- Tickets URL: <http://ghc.haskell.org/trac/ghc/query?id=6132%2C7650%2C4176%2C8556> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0