Stripmining and Cleanup

The compiler automatically stripmines your loop and generates a cleanup loop. This means you do not need to unroll your loops, and, in most cases, this will also enable more vectorization.

Before Vectorization
i=0; while(i<n) { // Original loop code a[i]=b[i]+c[i]; ++i; }

Before Vectorization

i=0;

while(i<n)

{

// Original loop code

a[i]=b[i]+c[i];

++i;

}

After Vectorization
// The vectorizer generates the following two loops i=0; while(i<(n-n%4)) { // Vector strip-mined loop // Subscript [i:i+3] denotes SIMD execution a[i:i+3]=b[i:i+3]+c[i:i+3]; i=i+4; } while(i<n) { // Scalar clean-up loop a[i]=b[i]+c[i]; }

After Vectorization

// The vectorizer generates the following two loops

i=0;

while(i<(n-n%4))

{

// Vector strip-mined loop

// Subscript [i:i+3] denotes SIMD execution

a[i:i+3]=b[i:i+3]+c[i:i+3];

i=i+4;

}

while(i<n)

{

// Scalar clean-up loop

a[i]=b[i]+c[i];

}