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
|
/*
* Simd Library (http://ermig1979.github.io/Simd).
*
* Copyright (c) 2011-2020 Yermalayeu Ihar.
*
* Permission is hereby granted, free of charge, to any person obtaining a copy
* of this software and associated documentation files (the "Software"), to deal
* in the Software without restriction, including without limitation the rights
* to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
* copies of the Software, and to permit persons to whom the Software is
* furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#include "Simd/SimdStore.h"
#include "Simd/SimdMemory.h"
namespace Simd
{
#ifdef SIMD_NEON_ENABLE
namespace Neon
{
const size_t A3 = A * 3;
const size_t A4 = A * 4;
union Bgra
{
uint8x16x4_t bgra;
uint8x16x3_t bgr;
};
template <bool align> SIMD_INLINE void BgrToBgra(const uint8_t * bgr, uint8_t * bgra, Bgra & _bgra)
{
_bgra.bgr = Load3<align>(bgr);
Store4<align>(bgra, _bgra.bgra);
}
template <bool align> void BgrToBgra(const uint8_t * bgr, size_t width, size_t height, size_t bgrStride, uint8_t * bgra, size_t bgraStride, uint8_t alpha)
{
assert(width >= A);
if (align)
assert(Aligned(bgra) && Aligned(bgraStride) && Aligned(bgr) && Aligned(bgrStride));
size_t alignedWidth = AlignLo(width, A);
Bgra _bgra;
_bgra.bgra.val[3] = vdupq_n_u8(alpha);
for (size_t row = 0; row < height; ++row)
{
for (size_t col = 0, colBgra = 0, colBgr = 0; col < alignedWidth; col += A, colBgra += A4, colBgr += A3)
BgrToBgra<align>(bgr + colBgr, bgra + colBgra, _bgra);
if (width != alignedWidth)
BgrToBgra<false>(bgr + 3 * (width - A), bgra + 4 * (width - A), _bgra);
bgr += bgrStride;
bgra += bgraStride;
}
}
void BgrToBgra(const uint8_t * bgr, size_t width, size_t height, size_t bgrStride, uint8_t * bgra, size_t bgraStride, uint8_t alpha)
{
if (Aligned(bgra) && Aligned(bgraStride) && Aligned(bgr) && Aligned(bgrStride))
BgrToBgra<true>(bgr, width, height, bgrStride, bgra, bgraStride, alpha);
else
BgrToBgra<false>(bgr, width, height, bgrStride, bgra, bgraStride, alpha);
}
//---------------------------------------------------------------------
template <bool align> SIMD_INLINE void Bgr48pToBgra32(uint8_t * bgra,
const uint8_t * blue, const uint8_t * green, const uint8_t * red, size_t offset, const uint8x16_t & alpha)
{
uint8x16x2_t _blue = Load2<align>(blue + offset);
uint8x16x2_t _green = Load2<align>(green + offset);
uint8x16x2_t _red = Load2<align>(red + offset);
uint8x16x4_t _bgra;
_bgra.val[0] = _blue.val[0];
_bgra.val[1] = _green.val[0];
_bgra.val[2] = _red.val[0];
_bgra.val[3] = alpha;
Store4<align>(bgra, _bgra);
}
template <bool align> void Bgr48pToBgra32(const uint8_t * blue, size_t blueStride, size_t width, size_t height,
const uint8_t * green, size_t greenStride, const uint8_t * red, size_t redStride, uint8_t * bgra, size_t bgraStride, uint8_t alpha)
{
assert(width >= A);
if (align)
{
assert(Aligned(blue) && Aligned(blueStride));
assert(Aligned(green) && Aligned(greenStride));
assert(Aligned(red) && Aligned(redStride));
assert(Aligned(bgra) && Aligned(bgraStride));
}
uint8x16_t _alpha;
_alpha = vdupq_n_u8(alpha);
size_t alignedWidth = AlignLo(width, A) * 2;
for (size_t row = 0; row < height; ++row)
{
for (size_t srcOffset = 0, dstOffset = 0; srcOffset < alignedWidth; srcOffset += DA, dstOffset += QA)
Bgr48pToBgra32<align>(bgra + dstOffset, blue, green, red, srcOffset, _alpha);
if (width != alignedWidth)
Bgr48pToBgra32<false>(bgra + (width - A) * 4, blue, green, red, (width - A) * 2, _alpha);
blue += blueStride;
green += greenStride;
red += redStride;
bgra += bgraStride;
}
}
void Bgr48pToBgra32(const uint8_t * blue, size_t blueStride, size_t width, size_t height,
const uint8_t * green, size_t greenStride, const uint8_t * red, size_t redStride, uint8_t * bgra, size_t bgraStride, uint8_t alpha)
{
if (Aligned(blue) && Aligned(blueStride) && Aligned(green) && Aligned(greenStride) &&
Aligned(red) && Aligned(redStride) && Aligned(bgra) && Aligned(bgraStride))
Bgr48pToBgra32<true>(blue, blueStride, width, height, green, greenStride, red, redStride, bgra, bgraStride, alpha);
else
Bgr48pToBgra32<false>(blue, blueStride, width, height, green, greenStride, red, redStride, bgra, bgraStride, alpha);
}
//---------------------------------------------------------------------
template <bool align> SIMD_INLINE void RgbToBgra(const uint8_t* rgb, uint8_t* bgra, uint8x16_t alpha)
{
uint8x16x3_t _rgb = Load3<align>(rgb);
uint8x16x4_t _bgra;
_bgra.val[0] = _rgb.val[2];
_bgra.val[1] = _rgb.val[1];
_bgra.val[2] = _rgb.val[0];
_bgra.val[3] = alpha;
Store4<align>(bgra, _bgra);
}
template <bool align> void RgbToBgra(const uint8_t* rgb, size_t width, size_t height, size_t rgbStride, uint8_t* bgra, size_t bgraStride, uint8_t alpha)
{
assert(width >= A);
if (align)
assert(Aligned(bgra) && Aligned(bgraStride) && Aligned(rgb) && Aligned(rgbStride));
size_t alignedWidth = AlignLo(width, A);
uint8x16_t _alpha = vdupq_n_u8(alpha);
for (size_t row = 0; row < height; ++row)
{
for (size_t col = 0, colBgra = 0, colRgb = 0; col < alignedWidth; col += A, colBgra += A4, colRgb += A3)
RgbToBgra<align>(rgb + colRgb, bgra + colBgra, _alpha);
if (width != alignedWidth)
RgbToBgra<false>(rgb + 3 * (width - A), bgra + 4 * (width - A), _alpha);
rgb += rgbStride;
bgra += bgraStride;
}
}
void RgbToBgra(const uint8_t* rgb, size_t width, size_t height, size_t rgbStride, uint8_t* bgra, size_t bgraStride, uint8_t alpha)
{
if (Aligned(bgra) && Aligned(bgraStride) && Aligned(rgb) && Aligned(rgbStride))
RgbToBgra<true>(rgb, width, height, rgbStride, bgra, bgraStride, alpha);
else
RgbToBgra<false>(rgb, width, height, rgbStride, bgra, bgraStride, alpha);
}
}
#endif// SIMD_NEON_ENABLE
}
|
