surge-lipol 0.2.12-alpha.0

surge synthesizer -- linear interpolation struct
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
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189

crate::ix!();

impl crate::LipolPs {

    /// # Safety
    ///
    /// need to make sure we can access nquads blocks
    /// safely from src
    pub unsafe fn multiply_block_to<NQ>(
        &mut self,
        src: *mut f32,
        dst: *mut f32,
        nquads: NQ) 
        where <NQ as TryInto<usize>>::Error: Debug, NQ: TryInto<usize>
    {
        let nquads: usize = nquads.try_into().unwrap();

        lipol_ps_sse_block!(self, nquads, 2; 
            |i:usize, ref mut y1, ref mut y2, ref mut dy| {

                let src = src as *mut __m128;
                let dst = dst as *mut __m128;

                let a: __m128 = _mm_mul_ps(*src.add(i) , *y1);

                *dst.add(i) = a;

                *y1 = _mm_add_ps(*y1, *dy);

                let b: __m128 = _mm_mul_ps(*src.add(i + 1), *y2);

                *dst.add(i + 1) = b;

                *y2 = _mm_add_ps(*y2, *dy);
            }
        );
    }

    /// # Safety
    ///
    /// need to make sure we can access nquads blocks
    /// safely from src
    pub unsafe fn multiply_2_blocks_to<NQ>(
        &mut self,
        src1: *mut f32,
        src2: *mut f32,
        dst1: *mut f32,
        dst2: *mut f32,
        nquads: NQ) 
        where <NQ as TryInto<usize>>::Error: Debug, NQ: TryInto<usize>
    {
        let nquads: usize = nquads.try_into().unwrap();

        lipol_ps_sse_block!(self, nquads, 2; 
            |i:usize, ref mut y1, ref mut y2, ref mut dy| {

                let src1 = src1 as *mut __m128;
                let src2 = src2 as *mut __m128;
                let dst1 = dst1 as *mut __m128;
                let dst2 = dst2 as *mut __m128;

                *dst1.add(i) = _mm_mul_ps(*src1.add(i), *y1);
                *dst2.add(i) = _mm_mul_ps(*src2.add(i), *y1);

                *y1 = _mm_add_ps(*y1, *dy);

                *dst1.add(i + 1) = _mm_mul_ps(*src1.add(i + 1), *y2);
                *dst2.add(i + 1) = _mm_mul_ps(*src2.add(i + 1), *y2);

                *y2 = _mm_add_ps(*y2, *dy);
            }
        );
    }

    /// # Safety
    ///
    /// need to make sure we can access nquads blocks
    /// safely from src
    pub unsafe fn multiply_2_blocks<NQ: TryInto<usize>>(
        &mut self,
        src1: *mut f32, //__restrict
        src2: *mut f32, //__restrict
        nquads: NQ)
        where <NQ as TryInto<usize>>::Error: Debug
    {
        let nquads: usize = nquads.try_into().unwrap();

        lipol_ps_sse_block!(self, nquads, 2; 
            |i:usize, ref mut y1, ref mut y2, ref mut dy| {

                let src1 = src1 as *mut __m128;
                let src2 = src2 as *mut __m128;

                *src1.add(i) = _mm_mul_ps(*src1.add(i), *y1);
                *src2.add(i) = _mm_mul_ps(*src2.add(i), *y1);

                *y1 = _mm_add_ps(*y1, *dy);

                *src1.add(i + 1) = _mm_mul_ps(*src1.add(i + 1), *y2);
                *src2.add(i + 1) = _mm_mul_ps(*src2.add(i + 1), *y2);

                *y2 = _mm_add_ps(*y2, *dy);
            }
        );
    }

    /// # Safety
    ///
    /// need to make sure we can access nquads blocks
    /// safely from src
    pub unsafe fn multiply_block<NQ>(&mut self, 
        src: *mut f32, 
        nquads: NQ)
        where <NQ as TryInto<usize>>::Error: Debug, NQ: TryInto<usize>
    {
        let nquads: usize = nquads.try_into().unwrap();

        lipol_ps_sse_block!(self, nquads, 8; 
            |i:usize, ref mut y1, ref mut y2, ref mut dy| {

                let a: __m128 = _mm_mul_ps(
                    _mm_load_ps(src.add(i)), 
                    *y1
                );

                _mm_store_ps(
                    src.add(i), 
                    a);

                *y1 =  _mm_add_ps(*y1, *dy);

                let b: __m128 =  _mm_mul_ps(
                    _mm_load_ps(src.add(i + 4)), 
                    *y2
                );

                _mm_store_ps(
                    src.add(i + 4), 
                    b);

                *y2 = _mm_add_ps(*y2, *dy);
            }
            );
    }

    /// saturates the interpolator each step (for
    /// amp envelopes)
    ///
    /// # Safety
    ///
    /// need to make sure we can access nquads blocks
    /// safely from src
    pub unsafe fn multiply_block_sat1<NQ>(&mut self, 
        src: *mut f32, 
        nquads: NQ) 
        where <NQ as TryInto<usize>>::Error: Debug, NQ: TryInto<usize>
    {
        let nquads: usize = nquads.try_into().unwrap();

        let satv: __m128 =  _mm_set1_ps(1.0);

        lipol_ps_sse_block!(self, nquads, 8; 
            |i:usize, ref mut y1, ref mut y2, ref mut dy| {

                _mm_store_ps(
                    src.add(i), 
                    _mm_mul_ps(
                        _mm_load_ps(src.add(i)), 
                        *y1)
                );

                *y1 = _mm_min_ps(
                    satv, 
                    _mm_add_ps(*y1, *dy)
                );

                _mm_store_ps(
                    src.add(i + 4), 
                    _mm_mul_ps(
                        _mm_load_ps(src.add(i + 4)), *y2)
                );

                *y2 = _mm_min_ps(
                    satv, 
                    _mm_add_ps(*y2, *dy)
                );
            });
    }
}