Question

I’m writing a game in Haskell, and my current pass at the UI involves a lot of procedural generation of geometry. I am currently focused on identifying performance of one particular operation (C-ish pseudocode):

Vec4f multiplier, addend;
Vec4f vecList[];
for (int i = 0; i < count; i++)
    vecList[i] = vecList[i] * multiplier + addend;

That is, a bog-standard multiply-add of four floats, the kind of thing ripe for SIMD optimization.

The result is going to an OpenGL vertex buffer, so it has to get dumped into a flat C array eventually. For the same reason, the calculations should probably be done on C ‘float’ types.

I’ve looked for either a library or a native idiomatic solution to do this sort of thing quickly in Haskell, but every solution I’ve come up with seems to hover around 2% of the performance (that is, 50x slower) compared to C from GCC with the right flags. Granted, I started with Haskell a couple weeks ago, so my experience is limited—which is why I’m coming to you guys. Can any of you offer suggestions for a faster Haskell implementation, or pointers to documentation on how to write high-performance Haskell code?

First, the most recent Haskell solution (clocks about 12 seconds). I tried the bang-patterns stuff from this SO post, but it didn’t make a difference AFAICT. Replacing ‘multAdd’ with ‘(\i v -> v * 4)’ brought execution time down to 1.9 seconds, so the bitwise stuff (and consequent challenges to automatic optimization) doesn’t seem to be too much at fault.

{-# LANGUAGE BangPatterns #-}
{-# OPTIONS_GHC -O2 -fvia-C -optc-O3 -fexcess-precision -optc-march=native #-}

import Data.Vector.Storable
import qualified Data.Vector.Storable as V
import Foreign.C.Types
import Data.Bits

repCount = 10000
arraySize = 20000

a = fromList $ [0.2::CFloat,  0.1, 0.6, 1.0]
m = fromList $ [0.99::CFloat, 0.7, 0.8, 0.6]

multAdd :: Int -> CFloat -> CFloat
multAdd !i !v = v * (m ! (i .&. 3)) + (a ! (i .&. 3))

multList :: Int -> Vector CFloat -> Vector CFloat
multList !count !src
    | count <= 0    = src
    | otherwise     = multList (count-1) $ V.imap multAdd src

main = do
    print $ Data.Vector.Storable.sum $ multList repCount $ 
        Data.Vector.Storable.replicate (arraySize*4) (0::CFloat)

Here’s what I have in C. The code here has a few #ifdefs which prevents it from being compiled straight-up; scroll down for the test driver.

#include <stdio.h>
#include <stdlib.h>
#include <time.h>

typedef float v4fs __attribute__ ((vector_size (16)));
typedef struct { float x, y, z, w; } Vector4;

void setv4(v4fs *v, float x, float y, float z, float w) {
    float *a = (float*) v;
    a[0] = x;
    a[1] = y;
    a[2] = z;
    a[3] = w;
}

float sumv4(v4fs *v) {
    float *a = (float*) v;
    return a[0] + a[1] + a[2] + a[3];
}

void vecmult(v4fs *MAYBE_RESTRICT s, v4fs *MAYBE_RESTRICT d, v4fs a, v4fs m) {
    for (int j = 0; j < N; j++) {
        d[j] = s[j] * m + a;
    }
}

void scamult(float *MAYBE_RESTRICT s, float *MAYBE_RESTRICT d,
             Vector4 a, Vector4 m) {
    for (int j = 0; j < (N*4); j+=4) {
        d[j+0] = s[j+0] * m.x + a.x;
        d[j+1] = s[j+1] * m.y + a.y;
        d[j+2] = s[j+2] * m.z + a.z;
        d[j+3] = s[j+3] * m.w + a.w;
    }
}

int main () {
    v4fs a, m;
    v4fs *s, *d;

    setv4(&a, 0.2, 0.1, 0.6, 1.0);
    setv4(&m, 0.99, 0.7, 0.8, 0.6);

    s = calloc(N, sizeof(v4fs));
    d = s;

    double start = clock();
    for (int i = 0; i < M; i++) {

#ifdef COPY
        d = malloc(N * sizeof(v4fs));
#endif

#ifdef VECTOR
        vecmult(s, d, a, m);
#else
        Vector4 aa = *(Vector4*)(&a);
        Vector4 mm = *(Vector4*)(&m);
        scamult((float*)s, (float*)d, aa, mm);
#endif

#ifdef COPY
        free(s);
        s = d;
#endif
    }
    double end = clock();

    float sum = 0;
    for (int j = 0; j < N; j++) {
        sum += sumv4(s+j);
    }
    printf("%-50s %2.5f %f\n\n", NAME,
            (end - start) / (double) CLOCKS_PER_SEC, sum);
}

This script will compile and run the tests with a number of gcc flag combinations. The best performance was had by cmath-64-native-O3-restrict-vector-nocopy on my system, taking 0.22 seconds.

import System.Process
import GHC.IOBase

cBase = ("cmath", "gcc mult.c -ggdb --std=c99 -DM=10000 -DN=20000")
cOptions = [
            [("32", "-m32"), ("64", "-m64")],
            [("generic", ""), ("native", "-march=native -msse4")],
            [("O1", "-O1"), ("O2", "-O2"), ("O3", "-O3")],
            [("restrict", "-DMAYBE_RESTRICT=__restrict__"),
                ("norestrict", "-DMAYBE_RESTRICT=")],
            [("vector", "-DVECTOR"), ("scalar", "")],
            [("copy", "-DCOPY"), ("nocopy", "")]
           ]

-- Fold over the Cartesian product of the double list. Probably a Prelude function
-- or two that does this, but hey. The 'perm' referred to permutations until I realized
-- that this wasn't actually doing permutations. '
permfold :: (a -> a -> a) -> a -> [[a]] -> [a]
permfold f z [] = [z]
permfold f z (x:xs) = concat $ map (\a -> (permfold f (f z a) xs)) x

prepCmd :: (String, String) -> (String, String) -> (String, String)
prepCmd (name, cmd) (namea, cmda) =
    (name ++ "-" ++ namea, cmd ++ " " ++ cmda)

runCCmd name compileCmd = do
    res <- system (compileCmd ++ " -DNAME=\\\"" ++ name ++ "\\\" -o " ++ name)
    if res == ExitSuccess
        then do system ("./" ++ name)
                return ()
        else    putStrLn $ name ++ " did not compile"

main = do
    mapM_ (uncurry runCCmd) $ permfold prepCmd cBase cOptions

Answer 1

Roman Leschinkskiy responds:

Actually, the core looks mostly ok to
me. Using unsafeIndex instead of (!)
makes the program more than twice as
fast (see my answer above). The
program below is much faster, though
(and cleaner, IMO). I suspect the
remaining difference between this and
the C program is due to GHC’s general
suckiness when it comes to floating
point. The HEAD produces the
best results with the NCG and -msse2

First, define a new Vec4 data type:

{-# LANGUAGE BangPatterns #-}

import Data.Vector.Storable
import qualified Data.Vector.Storable as V
import Foreign
import Foreign.C.Types

-- Define a 4 element vector type
data Vec4 = Vec4 {-# UNPACK #-} !CFloat
                 {-# UNPACK #-} !CFloat
                 {-# UNPACK #-} !CFloat
                 {-# UNPACK #-} !CFloat

Ensure we can store it in an array

instance Storable Vec4 where
  sizeOf _ = sizeOf (undefined :: CFloat) * 4
  alignment _ = alignment (undefined :: CFloat)

  {-# INLINE peek #-}
  peek p = do
             a <- peekElemOff q 0
             b <- peekElemOff q 1
             c <- peekElemOff q 2
             d <- peekElemOff q 3
             return (Vec4 a b c d)
    where
      q = castPtr p
  {-# INLINE poke #-}
  poke p (Vec4 a b c d) = do
             pokeElemOff q 0 a
             pokeElemOff q 1 b
             pokeElemOff q 2 c
             pokeElemOff q 3 d
    where
      q = castPtr p

Values and methods on this type:

a = Vec4 0.2 0.1 0.6 1.0
m = Vec4 0.99 0.7 0.8 0.6

add :: Vec4 -> Vec4 -> Vec4
{-# INLINE add #-}
add (Vec4 a b c d) (Vec4 a' b' c' d') = Vec4 (a+a') (b+b') (c+c') (d+d')

mult :: Vec4 -> Vec4 -> Vec4
{-# INLINE mult #-}
mult (Vec4 a b c d) (Vec4 a' b' c' d') = Vec4 (a*a') (b*b') (c*c') (d*d')

vsum :: Vec4 -> CFloat
{-# INLINE vsum #-}
vsum (Vec4 a b c d) = a+b+c+d

multList :: Int -> Vector Vec4 -> Vector Vec4
multList !count !src
    | count <= 0    = src
    | otherwise     = multList (count-1) $ V.map (\v -> add (mult v m) a) src

main = do
    print $ Data.Vector.Storable.sum
          $ Data.Vector.Storable.map vsum
          $ multList repCount
          $ Data.Vector.Storable.replicate arraySize (Vec4 0 0 0 0)

repCount, arraySize :: Int
repCount = 10000
arraySize = 20000

With ghc 6.12.1, -O2 -fasm:

• 1.752

With ghc HEAD (june 26), -O2 -fasm -msse2

• 1.708

This looks like the most idiomatic way to write a Vec4 array, and gets the best performance (11x faster than your original). (And this might become a benchmark for GHC’s LLVM backend)