So, I decided to try to write an efficient Lisp version of your C
code, but the C code is broken.
#include <stdio.h>
const int n = 1000;
const int nn = 1000*1000;
static double mat[1000*1000];
int main ()
{
int i, pos1, pos2;
for (i=0; i<nn; i++)
mat[i] = 1.0;
for (i=0; i<10; i++)
for (pos1=1; pos1<n-1; pos1++)
for (pos2=pos1+n; pos2<pos1+(n-1)*n; pos2+=n)
mat[pos2] = 0.11111111111111*
(mat[pos2-1001]+mat[pos2-1]+mat[pos2+999]+
mat[pos2-1000]+mat[pos2]+mat[pos2+1000]+
mat[pos2-999]+mat[pos2+1]+mat[pos2+1001]);
printf("%f", mat[n+1]);
return 0;
}
The first time through the nested for loops, you try to set mat[2] to the value
0.11111111111111 *
mat[2-1001] + mat[2-1] + mat[2+999] +
mat[2-1000] + mat[2] + mat[2+1000] +
mat[2-999] + mat[2+1] + mat[2+1001]
So, while there must be something accessible at memory locations
(&mat)-(sizeof(double)*999),
(&mat)-(sizeof(double)*998), and
(&mat)-(sizeof(double)*997)
whatever result you're getting is garbage. So, before you go
comparing the speed of the C and Lisp programs, you might want to make
sure the C program at least makes sense ;-)
--
/|_ .-----------------------.
,' .\ / | No to Imperialist war |
,--' _,' | Wage class war! |
/ / `-----------------------'
( -. |
| ) |
(`-. '--.)
`. )----'
