- Glasgow-haskell-users

RE: FFI
by Simon Marlow 12 Aug '02

12 Aug '02

> I have one program in haskell that function in C passes two > arrays for one > which adds the two and returns one 3º. How I make to import > this function > and to use it in my program? Which is the mapping of the > arrays of haskell > for the C? Perhaps you should take a look at the FFI spec, here: http://www.cse.unsw.edu.au/~chak/haskell/ffi/ and the information specific to using GHC with the FFI, here: http://www.haskell.org/ghc/docs/latest/html/users_guide/ffi.html Cheers, Simon

1 0

secret
by pretty 12 Aug '02

12 Aug '02

http://ragazzepericolose.da.ru veramente................ http://ragazzepericolose.da.ru

1 0

HOpenGL quo vadis?
by Sven Moritz Hallberg 10 Aug '02

10 Aug '02

GHC 5.04 builds the incomplete port of HOpenGL to the new hierarchical library structure. Why not the old but working hslibs version? Sven Moritz

2 2

withStorableArray making IORefs
by Hal Daume III 09 Aug '02

09 Aug '02

I'm guessing it's not safe to use "newIORef" as the function argument to withStorableArray, but in a simple case [1], it seems to work okay. I suppose the danger is if the RTS decides to move the array. Does this happen IRL? - Hal [1]: the example program is something like: module Main where import Data.Array.Storable import Foreign import Data.IORef mkArray :: Int -> IO (StorableArray Int Int) mkArray sz = newArray_ (1,sz) >>= \arr -> mapM_ (\i -> writeArray arr i i) [1..sz] >> return arr foo :: Ptr Int -> IO Int foo ptr = peek ptr >>= \v -> poke ptr (v+1) >> return v bar = do arr <- mkArray 10 ptr <- withStorableArray arr newIORef >>= readIORef foo ptr >> foo ptr >> foo ptr withStorableArray arr foo main = bar >>= print >> bar >>= print >> bar >>= print -- Hal Daume III "Computer science is no more about computers | hdaume(a)isi.edu than astronomy is about telescopes." -Dijkstra | www.isi.edu/~hdaume

1 0

Your message to Haskell awaits moderator approval
by haskell-admin＠haskell.org 09 Aug '02

09 Aug '02

Your mail to 'Haskell' with the subject Japanese lass' sexy pictures Is being held until the list moderator can review it for approval. The reason it is being held: Message has a suspicious header Either the message will get posted to the list, or you will receive notification of the moderator's decision.

1 0

calling a seperate program and supplying command line args.
by Jason Smith 08 Aug '02

08 Aug '02

How to basically? I've not had much luck finding much in the way of doc's in ghc on how to do this?...basically I have a compiler in haskell and I want to call a parser which is written in Java. The parser is a .NET binary, it spits out a text file containing the parse tree which I then suck up and do the usual compiler things with. I need to supply the java parser command line args i.e the file to parse etc.. Thanks heaps Jason.

3 2

specializations on float
by Hal Daume III 08 Aug '02

08 Aug '02

the ghc docs suggest using Doubles instead of Floats because GHC specializes Double but not float...I want to use floats to save memory and to test this claim, i grepped through the libraries for the SPECIALIZE keyword and find that with the exception of a few complex routines (which I don't care about), floats are just as specialized as doubles...is this the case, or am I missing something? the results of grepping are: moussor:libraries/ grep "SPECIALI[SZ]E" `find . | grep 'hs$'` | cut -d':' -f2- | grep Float {-# SPECIALISE instance Num (Complex Float) #-} {-# SPECIALISE instance Fractional (Complex Float) #-} {-# SPECIALISE instance Floating (Complex Float) #-} {-# SPECIALISE instance Floating (Complex Double) #-} {-# SPECIALIZE properFraction :: Float -> (Int, Float) #-} {-# SPECIALIZE round :: Float -> Int #-} {-# SPECIALIZE ceiling :: Float -> Int #-} {-# SPECIALIZE floor :: Float -> Int #-} {-# SPECIALIZE properFraction :: Float -> (Integer, Float) #-} {-# SPECIALIZE truncate :: Float -> Integer #-} {-# SPECIALIZE round :: Float -> Integer #-} {-# SPECIALIZE ceiling :: Float -> Integer #-} {-# SPECIALIZE floor :: Float -> Integer #-} {-# SPECIALIZE showEFloat :: {-# SPECIALIZE showFFloat :: {-# SPECIALIZE showGFloat :: moussor:libraries/ grep "SPECIALI[SZ]E" `find . | grep 'hs$'` | cut -d':' -f2- | grep Double {-# SPECIALISE conjugate :: Complex Double -> Complex Double #-} {-# SPECIALISE mkPolar :: Double -> Double -> Complex Double #-} {-# SPECIALISE cis :: Double -> Complex Double #-} {-# SPECIALISE polar :: Complex Double -> (Double,Double) #-} {-# SPECIALISE magnitude :: Complex Double -> Double #-} {-# SPECIALISE phase :: Complex Double -> Double #-} {-# SPECIALISE instance Num (Complex Double) #-} {-# SPECIALISE instance Fractional (Complex Double) #-} {-# SPECIALISE instance Floating (Complex Double) #-} {-# SPECIALIZE properFraction :: Double -> (Int, Double) #-} {-# SPECIALIZE round :: Double -> Int #-} {-# SPECIALIZE ceiling :: Double -> Int #-} {-# SPECIALIZE floor :: Double -> Int #-} {-# SPECIALIZE properFraction :: Double -> (Integer, Double) #-} {-# SPECIALIZE truncate :: Double -> Integer #-} {-# SPECIALIZE round :: Double -> Integer #-} {-# SPECIALIZE ceiling :: Double -> Integer #-} {-# SPECIALIZE floor :: Double -> Integer #-} {-# SPECIALISE fromRat :: Rational -> Double, -- Hal Daume III "Computer science is no more about computers | hdaume(a)isi.edu than astronomy is about telescopes." -Dijkstra | www.isi.edu/~hdaume

1 0

FFI
by Allan Bruno 07 Aug '02

07 Aug '02

Hi, my name is Allan, i'm from Brazil, and I have a doubt. I have one program in haskell that function in C passes two arrays for one which adds the two and returns one 3º. How I make to import this function and to use it in my program? Which is the mapping of the arrays of haskell for the C? Thanks !

1 0

>2gb memory usage ==> segementation fault?
by Hal Daume III 06 Aug '02

06 Aug '02

Hi all, me again :) I keep trying to get my program to run through with all my data, but once it hits around 2 or 2 1/2 gb of memory allocated (according to top), it seg faults and emits a rather sizeable (2.5 gb) core dump. Rather than sending around my 5k line program, here's a simple one which exhibits the same problem: -- UseMemory.hs module Main where import Control.Monad import Control.Monad.ST import Data.Array.ST import System mkArrays :: Int -> Int -> ST s (STArray s Int (STArray s Int Int)) mkArrays src dst = newArray_ (0,src) >>= \arr -> mapM_ (\i -> newArray (0,dst) i >>= writeArray arr i) [0..src] >> return arr sumArrays :: STArray s Int (STArray s Int Int) -> ST s Int sumArrays arr = foldM (\s i -> readArray arr i >>= \arr' -> foldM (\s j -> readArray arr' j >>= return . (s+)) s [0..snd (bounds arr')]) 0 [0..snd (bounds arr)] main = do [ssize,dsize] <- getArgs print $ runST (mkArrays (read ssize) (read dsize) >>= sumArrays) I compile with -O2 -fvia-c, and then run: 9:52am enescu:TryMPI/ ./UseMemory 1 20000 +RTS -K128m 20001 9:53am enescu:TryMPI/ ./UseMemory 1 2000000 +RTS -K128m -A64m 2000001 9:54am enescu:TryMPI/ ./UseMemory 100 2000000 +RTS -K128m -A64m Segmentation fault 1858.28u 51.66s 35:26.02 89.8% (The -A is because if you don't increase the GC size, the program executes ridiculously slowly.) Unfortunately, this behavior is entirely predictable; no matter what arguments you use, once it gets to about 2.2 gb of memory usage it dies. (This is running on a solaris 2.8 machine with 8gb main memory, about 5 or 6gb free.) Could someone explain to me what's going on here? - Hal -- Hal Daume III "Computer science is no more about computers | hdaume(a)isi.edu than astronomy is about telescopes." -Dijkstra | www.isi.edu/~hdaume

1 0

Lazy bind...
by MR K P SCHUPKE 05 Aug '02

05 Aug '02

Is there any way to do a lazy bind operation, something like a <- $ getLine return (Constructor $ a) this works for computations in the IO monad :- a <- unsafeInterleaveIO getLine return (Constructor $ a) but I need to do this for a general monad M. Keean Schupke Department of Electrical & Electronic Engineering, Imperial College London.

5 8