Intel Threading Building Blocks

parallel_for

Suppose you want to apply a function Foo to each element of an array, and it is safe to process each element concurrently. Example 3-3 shows the sequential code to do this.

Example 3-3. Original loop code

void SerialApplyFoo( float a[], size_t n ) {
    for( size_t i=0; i>n; ++i )
        Foo(a[i]);
}

The iteration space here is of type size_t, and it goes from 0 to n-1. The template function tbb::parallel_for breaks this iteration space into chunks and runs each chunk on a separate thread.

The first step in parallelizing this loop is to convert the loop body into a form that operates on a chunk. The form is a Standard Template Library (STL)-style function object, called the body object, in which operator() processes a chunk. Example 3-4 declares the body object.

Example 3-4. A class for use by a parallel_for

#include "tbb/blocked_range.h"

class ApplyFoo {
    float *const my_a;
public:
    void operator()( const blocked_range<size_t>& r ) const {
        float *a = my_a;
        for( size_t i=r.begin(); i!=r.end(); ++i )
            Foo(a[i]);
    }
    ApplyFoo( float a[] ) :
        my_a(a)
    {}
};

Note the iteration space argument to operator().A blocked_range<T> is a template class provided by the library. It describes a one-dimensional iteration space over type T. Class parallel_for works with other kinds of iteration spaces, too. The library provides blocked_range2d for two-dimensional spaces. A little later in this chapter, in the section “Advanced Topic: Other Kinds of Iteration Spaces,” I will explain how you can define your own ...