If I'm reading your code correctly, you never initialize the contents of
x[][]. You just use whatever random bits are left over on the stack
after the execution of any library loading and/or static constructors
that precede the execution of main.
I don't think it is wise to do performance testing on doubles with
uncontrolled values. You may end up with some NaNs on INFs or other
intermediate results that have much worse performance than normal values.
dgorur wrote:
int main(){
double x[DATA_COUNT][INPUT_DIM];
