//---------------------------------------------------------------------------// // Copyright (c) 2013-2014 Kyle Lutz // // Distributed under the Boost Software License, Version 1.0 // See accompanying file LICENSE_1_0.txt or copy at // http://www.boost.org/LICENSE_1_0.txt // // See http://boostorg.github.com/compute for more information. //---------------------------------------------------------------------------// #ifndef BOOST_COMPUTE_DETAIL_WORK_SIZE_HPP #define BOOST_COMPUTE_DETAIL_WORK_SIZE_HPP #include namespace boost { namespace compute { namespace detail { // Given a total number of values (count), a number of values to // process per thread (vtp), and a number of threads to execute per // block (tpb), this function returns the global work size to be // passed to clEnqueueNDRangeKernel() for a 1D algorithm. inline size_t calculate_work_size(size_t count, size_t vpt, size_t tpb) { size_t work_size = static_cast(std::ceil(float(count) / vpt)); if(work_size % tpb != 0){ work_size += tpb - work_size % tpb; } return work_size; } } // end detail namespace } // end compute namespace } // end boost namespace #endif // BOOST_COMPUTE_DETAIL_WORK_SIZE_HPP