permutation iterator, not thoroughly checked yet

example/permutationIterator/CMakeLists.txt

@@ -0,0 +1,4 @@
+cmake_minimum_required(VERSION 3.18)
+set(_TARGET_NAME "example_permutation_iterator")
+alpaka_add_executable(${_TARGET_NAME} src/permutationIterator-main.cpp)
+target_link_libraries(${_TARGET_NAME} PUBLIC vikunja::internalvikunja)

example/permutationIterator/src/permutationIterator-main.cpp

@@ -0,0 +1,177 @@
+/* Copyright 2021 Hauke Mewes, Simeon Ehrig, Victor
+ *
+ * This file is part of vikunja.
+ *
+ * This Source Code Form is subject to the terms of the Mozilla Public
+ * License, v. 2.0. If a copy of the MPL was not distributed with this
+ * file, You can obtain one at http://mozilla.org/MPL/2.0/.
+ */
+
+#include <vikunja/reduce/reduce.hpp>
+#include <vikunja/mem/iterator/PermutationIterator.hpp>
+
+#include <alpaka/alpaka.hpp>
+
+#include <iostream>
+#include <list>
+
+int main()
+{
+    // Define the accelerator here. Must be one of the enabled accelerators.
+    //using TAcc = alpaka::AccGpuCudaRt<alpaka::DimInt<3u>, std::uint64_t>;
+    using TAcc = alpaka::AccCpuSerial<alpaka::DimInt<3u>, std::uint64_t>;
+
+    // Type of the data that will be reduced
+    using TRed = uint64_t;
+
+    // Alpaka index type
+    using Idx = alpaka::Idx<TAcc>;
+    // Alpaka dimension type
+    using Dim = alpaka::Dim<TAcc>;
+    // number of elements to reduce
+    const Idx n = static_cast<Idx>(3);
+
+    /*
+    // define device, platform, and queue types.
+    using DevAcc = alpaka::Dev<TAcc>;
+    using PltfAcc = alpaka::Pltf<DevAcc>;
+    // using QueueAcc = alpaka::test::queue::DefaultQueue<alpaka::Dev<TAcc>>;
+    using PltfHost = alpaka::PltfCpu;
+    using DevHost = alpaka::Dev<PltfHost>;
+    using QueueAcc = alpaka::Queue<TAcc, alpaka::Blocking>;
+
+    // Get the host device.
+    DevHost devHost(alpaka::getDevByIdx<PltfHost>(0u));
+    // Select a device to execute on.
+    DevAcc devAcc(alpaka::getDevByIdx<PltfAcc>(0u));
+    // Get a queue on the accelerator device.
+    QueueAcc queueAcc(devAcc);
+
+    typedef std::vector<TRed> element_range_type;
+    typedef std::vector<TRed> index_type;
+
+    // Use Lambda function for reduction
+    // auto sum = [] ALPAKA_FN_HOST_ACC(vikunja::mem::iterator::PermutationIterator<ValIterator, IdxIterator> i, vikunja::mem::iterator::PermutationIterator<ValIterator, IdxIterator> j) { return i + j; };
+    // auto doubleNum = [] ALPAKA_FN_HOST_ACC(TRed const i) { return 2 * i; };
+    std::cout << "Testing accelerator: " << alpaka::getAccName<TAcc>() << " with size: " << n << "\n"
+              << "Testing permutation iterator with value: 10\n";
+    //*/
+
+    std::vector<TRed> values{ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 };
+    std::vector<TRed> indices{ 5, 7, 10 };
+
+    // Create the permutation iterator
+    typedef vikunja::mem::iterator::PermutationIterator<element_range_type::iterator, index_type::iterator> permutation_type;
+
+    permutation_type permutationIter(values.begin(), indices.begin(), n);
+
+    std::cout << "Permutation result:\n";
+
+    auto it = permutationIter.begin();
+
+    // for (int i = 0; i < n; ++i)
+    // {
+    //     std::cout << indices[i] << "\t: " << it[i] << "\n";
+    // }
+
+    std::cout << "*++permutationIter: ";
+    std::cout << *++permutationIter;
+    std::cout << "\n*permutationIter: ";
+    std::cout << *permutationIter;
+    std::cout << "\n\n";
+
+    std::cout << "*permutationIter++: ";
+    std::cout << *permutationIter++;
+    std::cout << "\n*permutationIter: ";
+    std::cout << *permutationIter;
+    std::cout << "\n\n";
+
+    permutationIter += 6;
+    std::cout << "permutationIter += 6;\n"
+              << "*permutationIter: ";
+    std::cout << *permutationIter;
+    std::cout << "\n\n";
+
+    permutationIter -= 2;
+    std::cout << "permutationIter -= 2;\n"
+              << "*permutationIter: ";
+    std::cout << *permutationIter;
+    std::cout << "\n\n";
+
+    std::cout << "--permutationIter: ";
+    std::cout << *--permutationIter;
+    std::cout << "\n*permutationIter: ";
+    std::cout << *permutationIter;
+    std::cout << "\n\n";
+
+    std::cout << "permutationIter--: ";
+    std::cout << *permutationIter--;
+    std::cout << "\n*permutationIter: ";
+    std::cout << *permutationIter;
+    std::cout << "\n\n";
+
+    std::cout << "*(permutationIter + 2): ";
+    std::cout << *(permutationIter + 2);
+    std::cout << "\n\n";
+
+    std::cout << "*(permutationIter - 3): ";
+    std::cout << *(permutationIter - 3);
+    std::cout << "\n\n";
+
+    std::cout << "*permutationIter[0]: ";
+    std::cout << permutationIter[0];
+    std::cout << "\n";
+
+    std::cout << "*permutationIter[3]: ";
+    std::cout << permutationIter[3];
+    std::cout << "\n";
+
+    // REDUCE CALL:
+    // Takes the arguments: accelerator device, host device, accelerator queue, size of data, pointer-like to memory,
+    // reduce lambda.
+    // Idx reduceResult = vikunja::reduce::deviceReduce<TAcc,
+    //     vikunja::workdiv::BlockBasedPolicy<TAcc>,
+    //     vikunja::mem::iterator::MemAccessPolicy<TAcc>,
+    //     TFunc,
+    //     TInputIterator,
+    //     DevAcc,
+    //     DevHost,
+    //     QueueAcc,
+    //     Idx,
+    //     TOperator = vikunja::operators::BinaryOp<
+    //         TAcc,
+    //         TFunc,
+    //         typename std::iterator_traits<TInputIterator>,
+    //         typename std::iterator_traits<TInputIterator>>,
+    //         typename TRed = typename TOperator::TRed>(
+    // // Idx reduceResult = vikunja::reduce::deviceReduce<TAcc>(
+    //     devAcc, 
+    //     devHost, 
+    //     queueAcc, 
+    //     n, 
+    //     permutationIterBegin,
+    //     sum);
+
+    // // check reduce result
+    // auto expectedResult = n * 10;
+    // std::cout << "Expected reduce result: " << expectedResult << ", real result: " << reduceResult << "\n";
+
+    // TRANSFORM_REDUCE CALL:
+    // Takes the arguments: accelerator device, host device, accelerator queue, size of data, pointer-like to memory,
+    // transform lambda, reduce lambda.
+    // Idx transformReduceResult = vikunja::reduce::deviceTransformReduce<TAcc>(
+    //     devAcc,
+    //     devHost,
+    //     queueAcc,
+    //     n,
+    //     permutationIterBegin,
+    //     doubleNum,
+    //     sum);
+
+    // // check transform result
+    // auto expectedTransformReduce = expectedResult * 2;
+    // std::cout << "Expected transform_reduce result: " << expectedTransformReduce
+    //           << ", real result: " << transformReduceResult << "\n";
+
+    return 0;
+}