CODEWITHSUNDEEP
coptimizationvectorizationsse
Bob Fang
Bob Fang

Reputation: 7371

How to apply SSE to an array whose length is not guaranteed to be multiple of 4?

I am currently working with some function like below:

 void vadd(float * a, float * b, int n){
     for(int i = 0; i < n; i++){
         a[i] += b[i];
     }
 }

This loop essentially can be rewritten using SSE but my question is how to handle the few elements that are left out if n is not a multiple of 4?

Thanks a lot, Bob

Upvotes: 1

Views: 136

Answers (1)

Fred Foo
Fred Foo

Reputation: 363487

You can handle the last n % 4 elements with a separate loop:

void vadd(float *a, float *b, int n)
{
    int i = 0;
    for (; i < n - n % 4; i += 4) {
        a[i + 0] += b[i + 0];
        a[i + 1] += b[i + 1];
        a[i + 2] += b[i + 2];
        a[i + 3] += b[i + 3];
    }
    for (; i < n; i++) {
        a[i] += b[i]; 
    }
}

Upvotes: 5

Related Questions