Exceptions and Cancellation

Intel® Threading Building Blocks (Intel® TBB) supports exceptions and cancellation. When code inside an Intel® TBB algorithm throws an exception, the following steps generally occur:

  1. The exception is captured. Any further exceptions inside the algorithm are ignored.

  2. The algorithm is cancelled. Pending iterations are not executed. If there is Intel® TBB parallelism nested inside, the nested parallelism may also be cancelled as explained in Cancellation and Nested Parallelism.

  3. Once all parts of the algorithm stop, an exception is thrown on the thread that invoked the algorithm.

The exception thrown in step 3 might be the original exception, or might merely be a summary of type captured_exception. The latter usually occurs on current systems because propagating exceptions between threads requires support for the C++ std::exception_ptr functionality. As compilers evolve to support this functionality, future versions of Intel® TBB might throw the original exception. So be sure your code can catch either type of exception. The following example demonstrates exception handling.

#include "tbb/tbb.h"
#include <vector>
#include <iostream>
 
using namespace tbb;
using namespace std;
 
vector<int> Data;
 
struct Update {
    void operator()( const blocked_range<int>& r ) const {
        for( int i=r.begin(); i!=r.end(); ++i )
            Data.at(i) += 1;
    }
};
 
int main() {
    Data.resize(1000);
    try {
        parallel_for( blocked_range<int>(0, 2000), Update());
    } catch( captured_exception& ex ) {
       cout << "captured_exception: " << ex.what() << endl;
    } catch( out_of_range& ex ) {
       cout << "out_of_range: " << ex.what() << endl;
    }
    return 0;
}

The parallel_for attempts to iterate over 2000 elements of a vector with only 1000 elements. Hence the expression Data.at(i) sometimes throws an exception std::out_of_range during execution of the algorithm. When the exception happens, the algorithm is cancelled and an exception thrown at the call site to parallel_for.