Re: openMP gcc vs icc, erratic results with gcc

[Tom St Denis <tstdenis@ellipticsemi.com>]

diego sandoval wrote:
> int main(int argc, char *argv[])
> {
>   double start, stop; /* times of beginning and end of procedure */
>   double e, pi, factorial, product;
>   int i;
>
>       for (i = 1; i<num_steps; i++) {
>
> 	factorial *= i;
> 	e += 1.0/factorial;
>       }
>   
>       for (i = 0; i < num_steps*10; i++) {
> 	/* we want 1/1 - 1/3 + 1/5 - 1/7 etc.
> 	   therefore we count by fours (0, 4, 8, 12...) and take
>              1/(0+1) =  1/1
> 	   - 1/(0+3) = -1/3
>
>              1/(4+1) =  1/5
> 	   - 1/(4+3) = -1/7 and so on */
> 	pi += 1.0/(i*4.0 + 1.0);
> 	pi -= 1.0/(i*4.0 + 3.0);
>       }
>   

Aren't they both competing for i?  Admittedly I don't know how OpenMP 
does it's threading (I'd just use pthreads) but both are using "i".

Try bringing it inside the blocks and see if that changes anything.

Tom