rustsynth-sys 0.5.6

Low level bindings to VapourSynth
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
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

/*
* Copyright (c) 2012-2019 Fredrik Mellbin
*
* This file is part of VapourSynth.
*
* VapourSynth is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* VapourSynth is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with VapourSynth; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/

#include <stddef.h>
#include <stdint.h>
#include <emmintrin.h>

#define VS_TRANSPOSE_IMPL
#define BLOCK_WIDTH_BYTE 16
#define BLOCK_HEIGHT_BYTE 8
#define BLOCK_WIDTH_WORD 8
#define BLOCK_HEIGHT_WORD 8
#define BLOCK_WIDTH_DWORD 4
#define BLOCK_HEIGHT_DWORD 4
#include "../transpose.h"

static void transpose_block_byte(const uint8_t * VS_RESTRICT src, ptrdiff_t src_stride, uint8_t * VS_RESTRICT dst, ptrdiff_t dst_stride)
{
    __m128i row0 = _mm_load_si128((const __m128i *)ADD_OFFSET(src, 0 * src_stride));
    __m128i row1 = _mm_load_si128((const __m128i *)ADD_OFFSET(src, 1 * src_stride));
    __m128i row2 = _mm_load_si128((const __m128i *)ADD_OFFSET(src, 2 * src_stride));
    __m128i row3 = _mm_load_si128((const __m128i *)ADD_OFFSET(src, 3 * src_stride));
    __m128i row4 = _mm_load_si128((const __m128i *)ADD_OFFSET(src, 4 * src_stride));
    __m128i row5 = _mm_load_si128((const __m128i *)ADD_OFFSET(src, 5 * src_stride));
    __m128i row6 = _mm_load_si128((const __m128i *)ADD_OFFSET(src, 6 * src_stride));
    __m128i row7 = _mm_load_si128((const __m128i *)ADD_OFFSET(src, 7 * src_stride));

    __m128i t0, t1, t2, t3, t4, t5, t6, t7;
    __m128i tt0, tt1, tt2, tt3, tt4, tt5, tt6, tt7;

    row0 = _mm_shuffle_epi32(row0, _MM_SHUFFLE(3, 1, 2, 0));
    row1 = _mm_shuffle_epi32(row1, _MM_SHUFFLE(3, 1, 2, 0));
    row2 = _mm_shuffle_epi32(row2, _MM_SHUFFLE(3, 1, 2, 0));
    row3 = _mm_shuffle_epi32(row3, _MM_SHUFFLE(3, 1, 2, 0));
    row4 = _mm_shuffle_epi32(row4, _MM_SHUFFLE(3, 1, 2, 0));
    row5 = _mm_shuffle_epi32(row5, _MM_SHUFFLE(3, 1, 2, 0));
    row6 = _mm_shuffle_epi32(row6, _MM_SHUFFLE(3, 1, 2, 0));
    row7 = _mm_shuffle_epi32(row7, _MM_SHUFFLE(3, 1, 2, 0));

    t0 = _mm_unpacklo_epi8(row0, row1);
    t1 = _mm_unpacklo_epi8(row2, row3);
    t2 = _mm_unpacklo_epi8(row4, row5);
    t3 = _mm_unpacklo_epi8(row6, row7);
    t4 = _mm_unpackhi_epi8(row0, row1);
    t5 = _mm_unpackhi_epi8(row2, row3);
    t6 = _mm_unpackhi_epi8(row4, row5);
    t7 = _mm_unpackhi_epi8(row6, row7);

    tt0 = _mm_unpacklo_epi16(t0, t1);
    tt1 = _mm_unpackhi_epi16(t0, t1);
    tt2 = _mm_unpacklo_epi16(t2, t3);
    tt3 = _mm_unpackhi_epi16(t2, t3);
    tt4 = _mm_unpacklo_epi16(t4, t5);
    tt5 = _mm_unpackhi_epi16(t4, t5);
    tt6 = _mm_unpacklo_epi16(t6, t7);
    tt7 = _mm_unpackhi_epi16(t6, t7);

    row0 = _mm_unpacklo_epi32(tt0, tt2);
    row1 = _mm_unpackhi_epi32(tt0, tt2);
    row2 = _mm_unpacklo_epi32(tt1, tt3);
    row3 = _mm_unpackhi_epi32(tt1, tt3);
    row4 = _mm_unpacklo_epi32(tt4, tt6);
    row5 = _mm_unpackhi_epi32(tt4, tt6);
    row6 = _mm_unpacklo_epi32(tt5, tt7);
    row7 = _mm_unpackhi_epi32(tt5, tt7);

    _mm_storel_epi64((__m128i *)ADD_OFFSET(dst, 0 * dst_stride), row0);
    _mm_storel_epi64((__m128i *)ADD_OFFSET(dst, 2 * dst_stride), row1);
    _mm_storel_epi64((__m128i *)ADD_OFFSET(dst, 8 * dst_stride), row2);
    _mm_storel_epi64((__m128i *)ADD_OFFSET(dst, 10 * dst_stride), row3);
    _mm_storel_epi64((__m128i *)ADD_OFFSET(dst, 4 * dst_stride), row4);
    _mm_storel_epi64((__m128i *)ADD_OFFSET(dst, 6 * dst_stride), row5);
    _mm_storel_epi64((__m128i *)ADD_OFFSET(dst, 12 * dst_stride), row6);
    _mm_storel_epi64((__m128i *)ADD_OFFSET(dst, 14 * dst_stride), row7);

    _mm_storeh_pd((double *)ADD_OFFSET(dst, 1 * dst_stride), _mm_castsi128_pd(row0));
    _mm_storeh_pd((double *)ADD_OFFSET(dst, 3 * dst_stride), _mm_castsi128_pd(row1));
    _mm_storeh_pd((double *)ADD_OFFSET(dst, 9 * dst_stride), _mm_castsi128_pd(row2));
    _mm_storeh_pd((double *)ADD_OFFSET(dst, 11 * dst_stride), _mm_castsi128_pd(row3));
    _mm_storeh_pd((double *)ADD_OFFSET(dst, 5 * dst_stride), _mm_castsi128_pd(row4));
    _mm_storeh_pd((double *)ADD_OFFSET(dst, 7 * dst_stride), _mm_castsi128_pd(row5));
    _mm_storeh_pd((double *)ADD_OFFSET(dst, 13 * dst_stride), _mm_castsi128_pd(row6));
    _mm_storeh_pd((double *)ADD_OFFSET(dst, 15 * dst_stride), _mm_castsi128_pd(row7));
}

static void transpose_block_word(const uint16_t * VS_RESTRICT src, ptrdiff_t src_stride, uint16_t * VS_RESTRICT dst, ptrdiff_t dst_stride)
{
    __m128i row0 = _mm_load_si128((const __m128i *)ADD_OFFSET(src, 0 * src_stride));
    __m128i row1 = _mm_load_si128((const __m128i *)ADD_OFFSET(src, 1 * src_stride));
    __m128i row2 = _mm_load_si128((const __m128i *)ADD_OFFSET(src, 2 * src_stride));
    __m128i row3 = _mm_load_si128((const __m128i *)ADD_OFFSET(src, 3 * src_stride));
    __m128i row4 = _mm_load_si128((const __m128i *)ADD_OFFSET(src, 4 * src_stride));
    __m128i row5 = _mm_load_si128((const __m128i *)ADD_OFFSET(src, 5 * src_stride));
    __m128i row6 = _mm_load_si128((const __m128i *)ADD_OFFSET(src, 6 * src_stride));
    __m128i row7 = _mm_load_si128((const __m128i *)ADD_OFFSET(src, 7 * src_stride));

    __m128i t0, t1, t2, t3, t4, t5, t6, t7;
    __m128i tt0, tt1, tt2, tt3, tt4, tt5, tt6, tt7;

    t0 = _mm_unpacklo_epi16(row0, row1);
    t1 = _mm_unpacklo_epi16(row2, row3);
    t2 = _mm_unpacklo_epi16(row4, row5);
    t3 = _mm_unpacklo_epi16(row6, row7);
    t4 = _mm_unpackhi_epi16(row0, row1);
    t5 = _mm_unpackhi_epi16(row2, row3);
    t6 = _mm_unpackhi_epi16(row4, row5);
    t7 = _mm_unpackhi_epi16(row6, row7);

    tt0 = _mm_unpacklo_epi32(t0, t1);
    tt1 = _mm_unpackhi_epi32(t0, t1);
    tt2 = _mm_unpacklo_epi32(t2, t3);
    tt3 = _mm_unpackhi_epi32(t2, t3);
    tt4 = _mm_unpacklo_epi32(t4, t5);
    tt5 = _mm_unpackhi_epi32(t4, t5);
    tt6 = _mm_unpacklo_epi32(t6, t7);
    tt7 = _mm_unpackhi_epi32(t6, t7);

    row0 = _mm_unpacklo_epi64(tt0, tt2);
    row1 = _mm_unpackhi_epi64(tt0, tt2);
    row2 = _mm_unpacklo_epi64(tt1, tt3);
    row3 = _mm_unpackhi_epi64(tt1, tt3);
    row4 = _mm_unpacklo_epi64(tt4, tt6);
    row5 = _mm_unpackhi_epi64(tt4, tt6);
    row6 = _mm_unpacklo_epi64(tt5, tt7);
    row7 = _mm_unpackhi_epi64(tt5, tt7);

    _mm_store_si128((__m128i *)ADD_OFFSET(dst, 0 * dst_stride), row0);
    _mm_store_si128((__m128i *)ADD_OFFSET(dst, 1 * dst_stride), row1);
    _mm_store_si128((__m128i *)ADD_OFFSET(dst, 2 * dst_stride), row2);
    _mm_store_si128((__m128i *)ADD_OFFSET(dst, 3 * dst_stride), row3);
    _mm_store_si128((__m128i *)ADD_OFFSET(dst, 4 * dst_stride), row4);
    _mm_store_si128((__m128i *)ADD_OFFSET(dst, 5 * dst_stride), row5);
    _mm_store_si128((__m128i *)ADD_OFFSET(dst, 6 * dst_stride), row6);
    _mm_store_si128((__m128i *)ADD_OFFSET(dst, 7 * dst_stride), row7);
}

static void transpose_block_dword(const uint32_t * VS_RESTRICT src, ptrdiff_t src_stride, uint32_t * VS_RESTRICT dst, ptrdiff_t dst_stride)
{
    __m128 row0 = _mm_load_ps((const float *)ADD_OFFSET(src, 0 * src_stride));
    __m128 row1 = _mm_load_ps((const float *)ADD_OFFSET(src, 1 * src_stride));
    __m128 row2 = _mm_load_ps((const float *)ADD_OFFSET(src, 2 * src_stride));
    __m128 row3 = _mm_load_ps((const float *)ADD_OFFSET(src, 3 * src_stride));

    _MM_TRANSPOSE4_PS(row0, row1, row2, row3);

    _mm_store_ps((float *)ADD_OFFSET(dst, 0 * dst_stride), row0);
    _mm_store_ps((float *)ADD_OFFSET(dst, 1 * dst_stride), row1);
    _mm_store_ps((float *)ADD_OFFSET(dst, 2 * dst_stride), row2);
    _mm_store_ps((float *)ADD_OFFSET(dst, 3 * dst_stride), row3);
}

void vs_transpose_plane_byte_sse2(const void * VS_RESTRICT src, ptrdiff_t src_stride, void * VS_RESTRICT dst, ptrdiff_t dst_stride, unsigned width, unsigned height)
{
    transpose_plane_byte(src, src_stride, dst, dst_stride, width, height);
}

void vs_transpose_plane_word_sse2(const void * VS_RESTRICT src, ptrdiff_t src_stride, void * VS_RESTRICT dst, ptrdiff_t dst_stride, unsigned width, unsigned height)
{
    transpose_plane_word(src, src_stride, dst, dst_stride, width, height);
}

void vs_transpose_plane_dword_sse2(const void * VS_RESTRICT src, ptrdiff_t src_stride, void * VS_RESTRICT dst, ptrdiff_t dst_stride, unsigned width, unsigned height)
{
    transpose_plane_dword(src, src_stride, dst, dst_stride, width, height);
}