reduce_perf-inl.h 4.53 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109
/*
 * Copyright (c) 2014, Siemens AG. All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are met:
 *
 * 1. Redistributions of source code must retain the above copyright notice,
 * this list of conditions and the following disclaimer.
 *
 * 2. Redistributions in binary form must reproduce the above copyright notice,
 * this list of conditions and the following disclaimer in the documentation
 * and/or other materials provided with the distribution.
 *
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
 * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE
 * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
 * POSSIBILITY OF SUCH DAMAGE.
 */

#ifndef EMBB_ALGORITHMS_PERF_REDUCE_PERF_TBB_INL_H_
#define EMBB_ALGORITHMS_PERF_REDUCE_PERF_TBB_INL_H_

#include <embb/base/perf/call_args.h>
#include <reduce_perf.h>
#include <embb/base/memory_allocation.h>
#include <embb/algorithms/reduce.h>
#include <index_iterator.h>
#include <embb/tasks/tasks.h>

namespace embb {
namespace algorithms {
namespace perf {

using embb::base::perf::CallArgs;

template<typename T>
SerialReduce<T>::SerialReduce(
  const embb::base::perf::CallArgs & args) :
  cargs(args),
  vector_size(args.VectorSize()) {
  if (cargs.StressMode() == CallArgs::RAM_STRESS) {
    v = static_cast<T*>(
      embb::base::Allocation::AllocateCacheAligned(
      vector_size * sizeof(T)));
    for (size_t i = 0; i < vector_size; i++) {
      v[i] = (T)i;
    }
  }
  else {
    v = 0;
  }
}

template<typename T>
SerialReduce<T>::~SerialReduce() {
  if (v != 0) {
    embb::base::Allocation::FreeAligned(v);
  }
}

template<typename T>
void SerialReduce<T>::Run() {
  result = 0;
  if (cargs.StressMode() == CallArgs::CPU_STRESS) {
    TransformOp<T> op(static_cast<T>(1.0) / vector_size, cargs);
    for (size_t i = 0; i < vector_size; i++) {
      result += op((T)i);
    }
    result /= static_cast<T>(vector_size);
  }
  else if (cargs.StressMode() == CallArgs::RAM_STRESS) {
    TransformOp<T> op(static_cast<T>(1.0) / vector_size, cargs);
    for (size_t i = 0; i < vector_size; i++) {
      result += op(v[i]);
    }
    result /= static_cast<T>(vector_size);
  }
}

template<typename T>
ParallelReduce<T>::ParallelReduce(
  const embb::base::perf::CallArgs & args) :
  cargs(args),
  vector_size(args.VectorSize()) {
  if (cargs.StressMode() == CallArgs::RAM_STRESS) {
    v = static_cast<T*>(
      embb::base::Allocation::AllocateCacheAligned(
      vector_size * sizeof(T)));
  }
  else {
    v = 0;
  }
}

template<typename T>
ParallelReduce<T>::~ParallelReduce() {
  if (v != 0) {
    embb::base::Allocation::FreeAligned(v);
  }
}

template<typename T>
110 111 112 113 114 115 116 117 118 119
void ParallelReduce<T>::Pre() {
  if (cargs.StressMode() == CallArgs::RAM_STRESS) {
    // Initialize input vector with incrementing values:
    for (size_t i = 0; i < vector_size; i++) {
      v[i] = (T)i;
    }
  }
}

template<typename T>
120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152
void ParallelReduce<T>::Run(unsigned int numThreads) {
  TransformOp<T> op(static_cast<T>(1) / vector_size, cargs);
  if (cargs.StressMode() == CallArgs::CPU_STRESS) {
    result = embb::algorithms::Reduce(
      // Using iterator returning index value to avoid 
      // memory access
      IndexIterator<T>(0),
      IndexIterator<T>(static_cast<int>(vector_size)),
      static_cast<T>(0),     // neutral element
      ::std::plus<T>(),      // reduce op
      op,                    // transform op
      embb::tasks::ExecutionPolicy(),
      vector_size / numThreads
      );
  }
  else if (cargs.StressMode() == CallArgs::RAM_STRESS) {
    result = embb::algorithms::Reduce(
      v, v + vector_size,    // input
      static_cast<T>(0),     // neutral element
      ::std::plus<T>(),      // reduce op
      op,                    // transform op
      embb::tasks::ExecutionPolicy(),
      vector_size / numThreads
      );
  }
  result /= static_cast<T>(vector_size);
}

} // namespace perf
} // namespace algorithms
} // namespace embb

#endif