parallel_iteration.h

Go to the documentation of this file.

/***************************************************************************
*   © 2012,2014 Advanced Micro Devices, Inc. All rights reserved.
*
*   Licensed under the Apache License, Version 2.0 (the "License");
*   you may not use this file except in compliance with the License.
*   You may obtain a copy of the License at
*
*       http://www.apache.org/licenses/LICENSE-2.0
*
*   Unless required by applicable law or agreed to in writing, software
*   distributed under the License is distributed on an "AS IS" BASIS,
*   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
*   See the License for the specific language governing permissions and
*   limitations under the License.

***************************************************************************/

#if !defined( BOLT_AMP_PARALLEL_ITER_H )
#define BOLT_AMP_PARALLEL_ITER_H

#pragma once

#include <amp.h>


//#include <bolt/combineable_queue.h>

//#define RESTRICT_AMP restrict(direct3d)


namespace bolt {

    template<typename IterType, typename Function>
    void parallel_iteration(concurrency::extent<1> ext, IterType Init, Function f)  {

        using namespace concurrency;

        parallel_for_each(extent<1>(ext), [=] (index<1> idx) mutable restrict(amp) {
            IterType iter(Init);

            bool keepGoing = true;

            while (keepGoing) {
                keepGoing = f(idx, iter);
            };
        });

    };

    template<typename IterType, typename Function>
    void parallel_iteration_1(concurrency::extent<1> ext, IterType Init, Function f)  {

        using namespace concurrency;
        int splitPoint = 3;  // Iterations to run on the GPU:

        static const int maxQSz=1024;
        array<pair<index<1>, IterType> outQueue(maxQSz);
        array<int,1>  qPtr(1);

        parallel_for_each(extent<1>(ext), [=] (index<1> idx) mutable restrict(amp) {
            IterType iter(Init);
            int iterations = splitPoint;

            bool keepGoing = true;
            while (keepGoing && --iterations) {
                keepGoing = f(idx, iter);
            };
            if (keepGoing) {
                // Enqueue for processing below - save idx and state.
                //; Horrible performance due to contended atomic operations, need to combine enqueue operations.
                int lPtr = atomic_fetch_add(&qPtr[0], 1);
                // FIXME - bounds check
                if (lPtr < maxQSz) {
                    outQueue[lPtr].first = idx;
                    outQueue[lPtr].second = iter;
                }
            };
        });
    };
};

// TODO: Need versions to handle more extent dimensions??

// FIXME-TODO
// Add Doxygen-style comments:

#endif