surge-math 0.2.12-alpha.0

surge synthesizer -- dsp math used in synthesis
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

crate::ix!();

#[cfg(target_arch = "x86_64")] #[inline] 
pub fn hardclip_ss(x: __m128) -> __m128
{
    unsafe {
        let x_min: __m128 = _mm_set_ss(-1.0);
        let x_max: __m128 = _mm_set_ss(1.0);
        _mm_max_ss(_mm_min_ss(x, x_max), x_min)
    }
}

#[cfg(target_arch = "x86_64")] #[inline] 
pub fn hardclip_ps(x: __m128) -> __m128
{
    unsafe {
        let x_min: __m128 = _mm_set1_ps(-1.0);
        let x_max: __m128 = _mm_set1_ps(1.0);
        _mm_max_ps(_mm_min_ps(x, x_max), x_min)
    }
}

#[cfg(target_arch = "x86_64")] #[inline] 
pub fn hardclip8_sd(x: __m128d) -> __m128d
{
    unsafe {
        let x_min: __m128d = _mm_set_sd(-7.0);
        let x_max: __m128d =  _mm_set_sd(8.0);
        _mm_max_sd(_mm_min_sd(x, x_max), x_min)
    }
}


///# Safety
///
///ensure we can access nquads * 4 elements
///contiguousely from a valid starting point x
pub unsafe fn hardclip_block<NQ>(x: *mut f32, nquads: NQ) 
    where <NQ as TryInto<usize>>::Error: Debug, NQ: TryInto<usize>
{
    let nquads: usize = nquads.try_into().unwrap();

    let x_min: __m128 = _mm_set1_ps(-1.0);
    let x_max: __m128 = _mm_set1_ps(1.0);

    let clip = |offset: usize| {
        let x = x.add(offset);
        let hi = _mm_min_ps( _mm_load_ps(x), x_max);
        let lo = _mm_max_ps( hi, x_min);
        _mm_store_ps(x, lo);
    };

    for i in (0_usize..(nquads << 2)).step_by(8)
    {
        clip(i);
        clip(i + 4);
    }
}

///# Safety
///
///ensure we can access nquads * 4 elements contiguousely 
///from a valid starting point x
pub unsafe fn hardclip_block8<NQ>(x: *mut f32, nquads: NQ) 
where 
    NQ: TryInto<usize>,
    <NQ as TryInto<usize>>::Error: Debug
{
    let nquads: usize = nquads.try_into().unwrap();

    let x_min: __m128 = _mm_set1_ps(-8.0);
    let x_max: __m128 = _mm_set1_ps(8.0);

    let clip = |offset: usize| {
        let x_in = x.add(offset);

        let elem = _mm_load_ps(x_in);

        let hi = _mm_min_ps(elem,x_max);

        let clipped = _mm_max_ps(hi, x_min);

        _mm_store_ps( x_in, clipped);
    };

    for i in (0_usize..(nquads << 2)).step_by(8) 
    {
        clip(i);
        clip(i + 4);
    }
}