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
/*************************************************************************
 *
 *  Project
 *                         _____ _____  __  __ _____
 *                        / ____|  __ \|  \/  |  __ \
 *  ___  _ __   ___ _ __ | |  __| |__) | \  / | |__) |
 * / _ \| '_ \ / _ \ '_ \| | |_ |  ___/| |\/| |  ___/
 *| (_) | |_) |  __/ | | | |__| | |    | |  | | |
 * \___/| .__/ \___|_| |_|\_____|_|    |_|  |_|_|
 *      | |
 *      |_|
 *
 * Copyright (C) Akiel Aries, <akiel@akiel.org>, et al.
 *
 * This software is licensed as described in the file LICENSE, which
 * you should have received as part of this distribution. The terms
 * among other details are referenced in the official documentation
 * seen here : https://akielaries.github.io/openGPMP/ along with
 * important files seen in this project.
 *
 * You may opt to use, copy, modify, merge, publish, distribute
 * and/or sell copies of the Software, and permit persons to whom
 * the Software is furnished to do so, under the terms of the
 * LICENSE file. As this is an Open Source effort, all implementations
 * must be of the same methodology.
 *
 *
 *
 * This software is distributed on an AS IS basis, WITHOUT
 * WARRANTY OF ANY KIND, either express or implied.
 *
 ************************************************************************/
#include <cassert>
#include <cstddef>
#include <cstdint>
#include <iostream>
#include <openGPMP/linalg/mtx.hpp>
#include <vector>

#if defined(__x86_64__) || defined(__amd64__) || defined(__amd64)

/************************************************************************
 *
 * Matrix Operations for AVX ISA
 *
 ************************************************************************/
#if defined(__AVX2__)

// AVX family intrinsics
#include <immintrin.h>

/************************************************************************
 *
 * Matrix Operations on Arrays
 *
 ************************************************************************/

// matrix addition using Intel intrinsics, accepts float arrays as matrices
void gpmp::linalg::Mtx::mtx_add(const float *A,
                                const float *B,
                                float *C,
                                int rows,
                                int cols) {
    if (rows > 16) {
        for (int i = 0; i < rows; ++i) {
            int j = 0;
            // requires at least size 4x4 size matrices
            for (; j < cols - 7; j += 8) {
                // load 4 elements from A, B, and C matrices using SIMD
                __m256 a = _mm256_loadu_ps(&A[i * cols + j]);
                __m256 b = _mm256_loadu_ps(&B[i * cols + j]);
                __m256 c = _mm256_loadu_ps(&C[i * cols + j]);<--- c is initialized
                // perform vectorized addition and accumulate the result
                c = _mm256_add_ps(a, b);<--- c is overwritten

                // store the result back to the C matrix
                _mm256_storeu_ps(&C[i * cols + j], c);
            }

            // handle the remaining elements that are not multiples of 8
            for (; j < cols; ++j) {
                C[i * cols + j] = A[i * cols + j] + B[i * cols + j];
            }
        }
    } else {
        // use standard matrix addition
        std_mtx_add(A, B, C, rows, cols);
    }
}

#endif

// x86
#endif