apply_large_kernel_to_input_AVX.h

Go to the documentation of this file.

/*
Created on Fri Jun 26 14:13:26 2020
Copyright 2020 Peter Rakyta, Ph.D.

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.

@author: Peter Rakyta, Ph.D.
*/
#ifndef apply_large_kernel_to_input_AVX_H
#define apply_large_kernel_to_input_AVX_H

#include <immintrin.h>
#include "matrix.h"
#include "matrix_float.h"
#include "common.h"

#include "apply_large_kernel_to_input.h"

//AVX auxiliary functions

__m256d get_AVX_vector(double* element_outer, double* element_inner);

__m256d complex_mult_AVX(__m256d input_vec, __m256d unitary_row_vec, __m256d neg);

//main function to call

void apply_large_kernel_to_input_AVX(Matrix& unitary, Matrix& input, std::vector<int> involved_qbits, const int& matrix_size);
void apply_large_kernel_from_right_AVX(Matrix& unitary, Matrix& input, std::vector<int> involved_qbits, const int& matrix_size);

void apply_large_kernel_to_input_AVX_OpenMP(Matrix& unitary, Matrix& input, std::vector<int> involved_qbits, const int& matrix_size);
void apply_large_kernel_from_right_AVX_OpenMP(Matrix& unitary, Matrix& input, std::vector<int> involved_qbits, const int& matrix_size);

void apply_large_kernel_to_input_AVX_TBB(Matrix& unitary, Matrix& input, std::vector<int> involved_qbits, const int& matrix_size);
void apply_large_kernel_from_right_AVX_TBB(Matrix& unitary, Matrix& input, std::vector<int> involved_qbits, const int& matrix_size);

// general N qubit kernel functions

inline void get_block_indices(int N, const std::vector<int> &tq, const std::vector<int> &non_targets,int iter_idx, std::vector<int> &indices);

inline void get_block_indices_fast(int iter_idx, const std::vector<int>& target_qubits, const std::vector<int>& non_targets, const std::vector<int>& block_pattern, std::vector<int>& indices);

void precompute_index_mapping(const std::vector<int>& target_qubits, const std::vector<int>& non_targets, std::vector<int>& block_pattern);

inline __m256d* construct_mv_xy_vectors(const Matrix& gate_kernel_unitary, const int& matrix_size);

void apply_2qbit_kernel_to_matrix_input_parallel_AVX_OpenMP(Matrix& two_qbit_unitary, Matrix& input, std::vector<int> involved_qbits, const int& matrix_size);


// CROT Kernels

void apply_crot_kernel_to_matrix_input_AVX_parallel(Matrix& u3_1qbit1,Matrix& u3_1qbit2,Matrix& input, const int& target_qbit, const int& control_qbit, const int& matrix_size);
void apply_crot_kernel_to_matrix_input_from_right_AVX_parallel(Matrix& u3_1qbit1,Matrix& u3_1qbit2,Matrix& input, const int& target_qbit, const int& control_qbit, const int& matrix_size);

void apply_crot_kernel_to_matrix_input_AVX(Matrix& u3_1qbit1, Matrix& u3_qbit2, Matrix& input, const int& target_qbit, const int& control_qbit, const int& matrix_size);
void apply_crot_kernel_to_matrix_input_from_right_AVX(Matrix& u3_1qbit1, Matrix& u3_qbit2, Matrix& input, const int& target_qbit, const int& control_qbit, const int& matrix_size);

void apply_large_kernel_to_input_AVX32(Matrix_float& unitary, Matrix_float& input, std::vector<int> involved_qbits, const int& matrix_size);
void apply_large_kernel_to_input_AVX_OpenMP32(Matrix_float& unitary, Matrix_float& input, std::vector<int> involved_qbits, const int& matrix_size);
void apply_large_kernel_to_input_AVX_TBB32(Matrix_float& unitary, Matrix_float& input, std::vector<int> involved_qbits, const int& matrix_size);
void apply_large_kernel_from_right_AVX32(Matrix_float& unitary, Matrix_float& input, std::vector<int> involved_qbits, const int& matrix_size);
void apply_large_kernel_from_right_AVX_OpenMP32(Matrix_float& unitary, Matrix_float& input, std::vector<int> involved_qbits, const int& matrix_size);
void apply_large_kernel_from_right_AVX_TBB32(Matrix_float& unitary, Matrix_float& input, std::vector<int> involved_qbits, const int& matrix_size);

void apply_2qbit_kernel_to_matrix_input_parallel_AVX_OpenMP32(Matrix_float& two_qbit_unitary, Matrix_float& input, std::vector<int> involved_qbits, const int& matrix_size);

void apply_crot_kernel_to_matrix_input_AVX_parallel32(Matrix_float& u3_1qbit1, Matrix_float& u3_1qbit2, Matrix_float& input, const int& target_qbit, const int& control_qbit, const int& matrix_size);
void apply_crot_kernel_to_matrix_input_AVX32(Matrix_float& u3_1qbit1, Matrix_float& u3_qbit2, Matrix_float& input, const int& target_qbit, const int& control_qbit, const int& matrix_size);
void apply_crot_kernel_to_matrix_input_from_right_AVX_parallel32(Matrix_float& u3_1qbit1, Matrix_float& u3_1qbit2, Matrix_float& input, const int& target_qbit, const int& control_qbit, const int& matrix_size);
void apply_crot_kernel_to_matrix_input_from_right_AVX32(Matrix_float& u3_1qbit1, Matrix_float& u3_qbit2, Matrix_float& input, const int& target_qbit, const int& control_qbit, const int& matrix_size);
#endif