image_swizzle 0.3.0

SIMD accelerated image swizzling routines
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

const SERIAL_BATCH_WIDTH: usize = 16;

macro_rules! swizzle_4_wide {
    ($src:expr, $dst:expr, $idxs:expr) => {
        assert!($src.len() % 4 == 0 && $src.len() == $dst.len() && $idxs.len() == 4);

        let end = 0;
        // TODO: Check if this is more performant
        // let end = ($src.len() / SERIAL_BATCH_WIDTH) * SERIAL_BATCH_WIDTH;
        // (0..end).step_by(SERIAL_BATCH_WIDTH).for_each(|i| {
        //     let vec = &$src[i..i + SERIAL_BATCH_WIDTH];
        //     #[rustfmt::skip]
        //     let swizzled = &[ // Hardcoded SERIAL_BATCH_WIDTH = 16
        //         vec[$idxs[0]], vec[$idxs[1]], vec[$idxs[2]], vec[$idxs[3]],
        //         vec[$idxs[0] + 4], vec[$idxs[1] + 4], vec[$idxs[2] + 4], vec[$idxs[3] + 4],
        //         vec[$idxs[0] + 8], vec[$idxs[1] + 8], vec[$idxs[2] + 8], vec[$idxs[3] + 8],
        //         vec[$idxs[0] + 12], vec[$idxs[1] + 12], vec[$idxs[2] + 12], vec[$idxs[3] + 12],
        //     ];
        //     $dst[i..i + SERIAL_BATCH_WIDTH].copy_from_slice(swizzled);
        // });

        (end..$src.len()).step_by(4).for_each(|i| {
            let (a, b, c, d) = ($src[i + 0], $src[i + 1], $src[i + 2], $src[i + 3]);
            $dst[i + $idxs[0]] = a;
            $dst[i + $idxs[1]] = b;
            $dst[i + $idxs[2]] = c;
            $dst[i + $idxs[3]] = d;
        });
    };
}

macro_rules! change_alpha_fourth {
    ($src:expr, $dst:expr, $to:expr) => {
        assert!($src.len() % 4 == 0 && $src.len() == $dst.len());
        (0..$src.len()).step_by(4).for_each(|i| {
            $dst[i] = $src[i];
            $dst[i + 1] = $src[i + 1];
            $dst[i + 2] = $src[i + 2];
            $dst[i + 3] = $to;
        })
    };
}

macro_rules! change_alpha_fourth_and_swizzle {
    ($src:expr, $dst:expr, $to:expr, $idxs:expr) => {
        assert!($src.len() % 4 == 0 && $src.len() == $dst.len() && $idxs.len() == 4);
        (0..$src.len()).step_by(4).for_each(|i| {
            let (a, b, c) = ($src[i + 0], $src[i + 1], $src[i + 2]);
            $dst[i + $idxs[0]] = a;
            $dst[i + $idxs[1]] = b;
            $dst[i + $idxs[2]] = c;
            $dst[i + $idxs[3]] = $to;
        })
    };
}

#[inline(always)]
pub fn rgba_to_bgra_inplace(src: &mut [u8]) {
    swizzle_4_wide!(src, src, [2, 1, 0, 3]);
}

#[inline(always)]
pub fn rgba_to_bgra(src: &[u8], dst: &mut [u8]) {
    swizzle_4_wide!(src, dst, [2, 1, 0, 3]);
}

#[inline(always)]
pub fn rgba32_to_bgra_inplace(src: &mut [u8]) {
    #[cfg(target_endian = "little")]
    swizzle_4_wide!(src, src, [1, 2, 3, 0]);
    #[cfg(target_endian = "big")]
    swizzle_4_wide!(src, src, [2, 1, 0, 3]);
}

#[inline(always)]
pub fn rgba32_to_rgba_inplace(src: &mut [u8]) {
    #[cfg(target_endian = "little")]
    swizzle_4_wide!(src, src, [3, 2, 1, 0]);
}

#[inline(always)]
pub fn bgra_to_rgba_inplace(src: &mut [u8]) {
    swizzle_4_wide!(src, src, [2, 1, 0, 3]);
}

#[inline(always)]
pub fn bgra_to_rgba(src: &[u8], dst: &mut [u8]) {
    swizzle_4_wide!(src, dst, [2, 1, 0, 3]);
}

#[inline(always)]
pub fn rgb0_to_rgbx_inplace(src: &mut [u8]) {
    change_alpha_fourth!(src, src, 255);
}

#[inline(always)]
pub fn rgb0_to_rgbx(src: &[u8], dst: &mut [u8]) {
    change_alpha_fourth!(src, dst, 255);
}

#[inline(always)]
pub fn bgr0_to_bgrx_inplace(src: &mut [u8]) {
    change_alpha_fourth!(src, src, 255);
}

#[inline(always)]
pub fn bgr0_to_bgrx(src: &[u8], dst: &mut [u8]) {
    change_alpha_fourth!(src, dst, 255);
}

#[inline(always)]
pub fn rgb0_to_bgrx_inplace(src: &mut [u8]) {
    change_alpha_fourth_and_swizzle!(src, src, 255, [2, 1, 0, 3]);
}

#[inline(always)]
pub fn rgb0_to_bgrx(src: &[u8], dst: &mut [u8]) {
    change_alpha_fourth_and_swizzle!(src, dst, 255, [2, 1, 0, 3]);
}

#[inline(always)]
pub fn bgr0_to_rgbx_inplace(src: &mut [u8]) {
    change_alpha_fourth_and_swizzle!(src, src, 255, [2, 1, 0, 3]);
}

#[inline(always)]
pub fn bgr0_to_rgbx(src: &[u8], dst: &mut [u8]) {
    change_alpha_fourth_and_swizzle!(src, dst, 255, [2, 1, 0, 3]);
}

#[inline(always)]
pub fn argb_to_rgba_inplace(src: &mut [u8]) {
    swizzle_4_wide!(src, src, [1, 2, 3, 0]);
}

#[cfg(test)]
mod tests {
    use super::*;

    crate::common::impl_tests!{}
}