scalet 0.1.2

Continious wavelet transform
Documentation 
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
110
111
112
113
114
115
116
117
118
119
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

/*
 * // Copyright (c) Radzivon Bartoshyk 5/2025. 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.
 * //
 * // 3.  Neither the name of the copyright holder nor the names of its
 * // contributors may be used to endorse or promote products derived from
 * // this software without specific prior written permission.
 * //
 * // 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.
 */
use crate::spetrum_arith::SpectrumArithmetic;
use num_complex::Complex;
use std::arch::x86_64::*;

// a * b.conj()
#[inline]
#[target_feature(enable = "avx2", enable = "fma")]
fn _mm256_fcmul_pd_conj_b(a: __m256d, b: __m256d) -> __m256d {
    // Swap real and imaginary parts of 'a' for FMA
    let a_yx = _mm256_shuffle_pd::<0b0101>(a, a); // [a_im, a_re, b_im, b_re]

    // Duplicate real and imaginary parts of 'b'
    let b_xx = _mm256_shuffle_pd::<0b0000>(b, b); // [c_re, c_re, d_re, d_re]
    let b_yy = _mm256_shuffle_pd::<0b1111>(b, b); // [c_im, c_im, d_im, d_im]

    _mm256_fmsubadd_pd(a, b_xx, _mm256_mul_pd(a_yx, b_yy))
}

// a * b.conj()
#[inline]
#[target_feature(enable = "avx2", enable = "fma")]
fn _mm_fcmul_pd_conj_b(a: __m128d, b: __m128d) -> __m128d {
    let temp1 = _mm_unpacklo_pd(b, b);
    let mut temp2 = _mm_unpackhi_pd(b, b);
    temp2 = _mm_mul_pd(temp2, a);
    temp2 = _mm_shuffle_pd::<0x01>(temp2, temp2);
    _mm_fmsubadd_pd(temp1, a, temp2)
}

#[derive(Copy, Clone, Default)]
pub(crate) struct AvxSpectrumF64 {}

impl SpectrumArithmetic<f64> for AvxSpectrumF64 {
    fn mul_by_b_conj_normalize(
        &self,
        dst: &mut [Complex<f64>],
        input: &[Complex<f64>],
        other: &[Complex<f64>],
        normalize_value: f64,
    ) {
        unsafe {
            self.mul_by_b_conj_normalize_impl(dst, input, other, normalize_value);
        }
    }
}

impl AvxSpectrumF64 {
    #[target_feature(enable = "avx2", enable = "fma")]
    fn mul_by_b_conj_normalize_impl(
        &self,
        dst: &mut [Complex<f64>],
        input: &[Complex<f64>],
        other: &[Complex<f64>],
        normalize_value: f64,
    ) {
        unsafe {
            let v_norm_factor = _mm256_set1_pd(normalize_value);

            for ((dst, input), other) in dst
                .chunks_exact_mut(8)
                .zip(input.chunks_exact(8))
                .zip(other.chunks_exact(8))
            {
                let vd0 = _mm256_loadu_pd(input.as_ptr().cast());
                let vd1 = _mm256_loadu_pd(input.get_unchecked(2..).as_ptr().cast());
                let vd2 = _mm256_loadu_pd(input.get_unchecked(4..).as_ptr().cast());
                let vd3 = _mm256_loadu_pd(input.get_unchecked(6..).as_ptr().cast());

                let vk0 = _mm256_loadu_pd(other.as_ptr().cast());
                let vk1 = _mm256_loadu_pd(other.get_unchecked(2..).as_ptr().cast());
                let vk2 = _mm256_loadu_pd(other.get_unchecked(4..).as_ptr().cast());
                let vk3 = _mm256_loadu_pd(other.get_unchecked(6..).as_ptr().cast());

                let d0 = _mm256_mul_pd(_mm256_fcmul_pd_conj_b(vd0, vk0), v_norm_factor);
                let d1 = _mm256_mul_pd(_mm256_fcmul_pd_conj_b(vd1, vk1), v_norm_factor);
                let d2 = _mm256_mul_pd(_mm256_fcmul_pd_conj_b(vd2, vk2), v_norm_factor);
                let d3 = _mm256_mul_pd(_mm256_fcmul_pd_conj_b(vd3, vk3), v_norm_factor);

                _mm256_storeu_pd(dst.as_mut_ptr().cast(), d0);
                _mm256_storeu_pd(dst.get_unchecked_mut(2..).as_mut_ptr().cast(), d1);
                _mm256_storeu_pd(dst.get_unchecked_mut(4..).as_mut_ptr().cast(), d2);
                _mm256_storeu_pd(dst.get_unchecked_mut(6..).as_mut_ptr().cast(), d3);
            }

            let dst_rem = dst.chunks_exact_mut(8).into_remainder();
            let input_rem = input.chunks_exact(8).remainder();
            let other_rem = other.chunks_exact(8).remainder();

            for ((dst, input), other) in dst_rem
                .chunks_exact_mut(8)
                .zip(input_rem.chunks_exact(8))
                .zip(other_rem.chunks_exact(8))
            {
                let a0 = _mm256_loadu_pd(input.as_ptr().cast());
                let b0 = _mm256_loadu_pd(other.as_ptr().cast());

                let d0 = _mm256_mul_pd(_mm256_fcmul_pd_conj_b(a0, b0), v_norm_factor);

                _mm256_storeu_pd(dst.as_mut_ptr().cast(), d0);
            }

            let dst_rem = dst_rem.chunks_exact_mut(8).into_remainder();
            let input_rem = input_rem.chunks_exact(8).remainder();
            let other_rem = other_rem.chunks_exact(8).remainder();

            for ((dst, input), other) in dst_rem
                .iter_mut()
                .zip(input_rem.iter())
                .zip(other_rem.iter())
            {
                let v0 = _mm_loadu_pd(input as *const Complex<f64> as *const _);
                let v1 = _mm_loadu_pd(other as *const Complex<f64> as *const _);

                let lo = _mm_mul_pd(
                    _mm_fcmul_pd_conj_b(v0, v1),
                    _mm256_castpd256_pd128(v_norm_factor),
                );

                _mm_storeu_pd(dst as *mut Complex<f64> as *mut _, lo);
            }
        }
    }
}