The TBB documentation gives this example of using lambda expressions with parallel_for, but doesn't provide an example of using lambda expressions with tbb::task::enqueue
.
I am looking for a simple example of tbb::task::enqueue
with a lambda expression.
Low-level tasks in TBB do not directly support lambda expressions. But with some extra coding you might create syntax-sugar helpers to do what you want.
You'd need to create a task class that calls a given functor:
template<typename F>
class lambda_task : public tbb::task {
F my_func;
/*override*/ tbb::task* execute() {
my_func();
return NULL;
}
public:
lambda_task( const F& f ) : my_func(f) {}
};
And then, you'd need to create a function template that takes a functor/lambda, wraps it into lambda_task
, and enqueues:
template<typename F>
void tbb_enqueue_lambda( const F& f ) {
tbb::task::enqueue( *new( tbb::task::allocate_root() ) lambda_task<F>(f) );
}
And then you might use this function with lambda expressions:
tbb_enqueue_lambda( []{ /* code here */ } );
The official TBB API classes that support lambda expressions, such as task_group
and task_arena
, use very similar code internally.
Update: to pass a function pointer and arguments to call it with, the above approach can be extended in some ways:
- In C++03, you'd need to add separate class templates for a task with one argument, two arguments, etc., and corresponding overloads of
tbb_enqueue_lambda
function - In C++11, you could use variadic templates, storing the actual arguments in an
std::tuple
insidelambda_task
, and 'unpacking' those for the function call. Unpacking is not trivial, but there are a few SO topics covering that already: "unpacking" a tuple to call a matching function pointer, How do I expand a tuple into variadic template function's arguments?, and other.
来源:https://stackoverflow.com/questions/22345935/threaded-building-blocks-tbb-enqueue-task-with-lambda