parallel_for_each¶
[algorithms.parallel_for_each]
Function template that processes work items in parallel.
// Defined in header <oneapi/tbb/parallel_for_each.h>
namespace oneapi {
namespace tbb {
template<typename InputIterator, typename Body>
void parallel_for_each( InputIterator first, InputIterator last, Body body );
template<typename InputIterator, typename Body>
void parallel_for_each( InputIterator first, InputIterator last, Body body, task_group_context& context );
template<typename Container, typename Body>
void parallel_for_each( Container& c, Body body );
template<typename Container, typename Body>
void parallel_for_each( Container& c, Body body, task_group_context& context );
template<typename Container, typename Body>
void parallel_for_each( const Container& c, Body body );
template<typename Container, typename Body>
void parallel_for_each( const Container& c, Body body, task_group_context& context );
} // namespace tbb
} // namespace oneapi
Requirements:
The
Bodytype must meet the ParallelForEachBody requirements.The
InputIteratortype must meet the Input Iterator requirements from the [input.iterators] ISO C++ Standard section.The
Containertype must meet the ContainerBasedSequence requirements.
The parallel_for_each template has two forms.
The sequence form parallel_for_each(first, last, body) applies a function object body over a
sequence [first,last). Items may be processed in parallel.
The container form parallel_for_each(c, body) is equivalent to parallel_for_each(std::begin(c), std::end(c), body).
All overloads can accept a task_group_context object so that the algorithm’s tasks are executed in this context. By default, the algorithm is executed in a bound context of its own.
feeder Class¶
Additional work items can be added by body if it has a second argument of type feeder.
The function terminates when body(x) returns for all items x that were in the input
sequence or added by method feeder::add.
Example¶
The following code sketches a body with the two-argument form of operator().
struct MyBody {
void operator()(item_t item, parallel_do_feeder<item_t>& feeder ) {
for each new piece of work implied by item do {
item_t new_item = initializer;
feeder.add(new_item);
}
}
};