Re: Helping out the Vectorizer

[Gordan Bobic <gordan@xxxxxxxxxx>]

On Tue, 14 Aug 2007, Daniel Berlin wrote:

> > I'm trying to get more of my code to vectorize, but very few of the loops
> > do. ICC (sorry for mentioning it, don't mean to offend anyone) manages to
> > vectorize all of the loops in my code, so it is doable (and succeeding
> > seems to yield a speed-up of 5-10x!).
> >
> > Is there a way to help out the vectorizer with pragmas and suchlike, to
> > resolve possible misperceived vector dependence issues?
>
> For starters, it would help to tell us which version of GCC you are 
> trying with.

4.1.0 and 4.2.1 both produce pretty much identical results.

> Second, your code doesn't compile as you've written it, so i can't
> diagnose more without you fixing it :)

OK, attached a compilable version. Compiler says:

test.cxx:15: note: not vectorized: can't determine dependence between 
this_10->D.2134.Curve[i.0_30] and this_10->D.2134.Curve[i.0_30]
test.cxx:23: note: not vectorized: unsupported use in stmt.
test.cxx:32: note: not vectorized: unsupported use in stmt.

> Third, even with some modifications to make it compile, your class
> isn't used, so we wouldn't bother to compile it.

It compiles now, as per the attached code file. I would have thought the 
code to demonstrate that a one liner loop doesn't vectorize should be 
enough to diagnose a problem with the vectorizer. I'm not asking for help 
with debugging my code, I'm asking for advice on how to make a fairly 
obviously vectorizable loop vectorize.

Thanks.

Gordan

#include "test.h"
float CurveFit::Foo ()
{
        static float BestFit[4];
	static unsigned int i;
	static unsigned int x;
	static float xx;
	static unsigned int LocalDataC;
	static float CurrentError;

	CurrentError = 0;
	LocalDataC = DataC;

	// This should vectorize but doesn't
        for (i = 0; i < 4; i++)
                Curve[i] += BestFit[i];

        // ...

	static float CacheX[1024];

        // This should vectorize but doesn't
        for (x = 0, xx = 0; x < LocalDataC; x++)
                CacheX[x] = a * xx++;

        // ...

        static float Temp;
        static float CacheParam[1024];

	// This should vectorize but doesn't
        for (x = 0; x < LocalDataC; x++)
        {
		Temp = DataV[x] - CacheParam[x] - Curve[3];
                CurrentError += (Temp * Temp);
        }

	// ...

	return CurrentError;
}

int main ()
{
	CurveFit *test;
	test = new CurveFit;
}

class CurveFit
{
	public:
		union
		{
			float Curve[4];
			struct
			{
				float a;
				float b;
				float c;
				float d;
			};
		};

		unsigned int    DataC;
		float           DataV[1024];

		float Foo ();
};