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
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
|
/*
* Simd Library (http://ermig1979.github.io/Simd).
*
* Copyright (c) 2011-2022 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/SimdMemory.h"
#include "Simd/SimdStore.h"
#include "Simd/SimdResizer.h"
#include "Simd/SimdResizerCommon.h"
#include "Simd/SimdSet.h"
#include "Simd/SimdUpdate.h"
#include "Simd/SimdUnpack.h"
namespace Simd
{
#ifdef SIMD_SSE41_ENABLE
namespace Sse41
{
ResizerByteArea1x1::ResizerByteArea1x1(const ResParam & param)
: Base::ResizerByteArea1x1(param)
{
}
SIMD_INLINE __m128i SaveLoadTail(const uint8_t * ptr, size_t tail)
{
uint8_t buffer[DA];
_mm_storeu_si128((__m128i*)(buffer), _mm_loadu_si128((__m128i*)(ptr + tail - A)));
return _mm_loadu_si128((__m128i*)(buffer + A - tail));
}
template<UpdateType update> SIMD_INLINE void ResizerByteArea1x1RowUpdate(const uint8_t * src0, size_t size, int32_t a, int32_t * dst)
{
if (update == UpdateAdd && a == 0)
return;
__m128i alpha = SetInt16(a, a);
size_t sizeA = AlignLo(size, A);
size_t i = 0;
for (; i < sizeA; i += A, dst += A)
{
__m128i s0 = _mm_loadu_si128((__m128i*)(src0 + i));
__m128i i0 = UnpackU8<0>(s0);
__m128i i1 = UnpackU8<1>(s0);
Update<update, true>(dst + 0 * F, _mm_madd_epi16(alpha, UnpackU8<0>(i0)));
Update<update, true>(dst + 1 * F, _mm_madd_epi16(alpha, UnpackU8<1>(i0)));
Update<update, true>(dst + 2 * F, _mm_madd_epi16(alpha, UnpackU8<0>(i1)));
Update<update, true>(dst + 3 * F, _mm_madd_epi16(alpha, UnpackU8<1>(i1)));
}
if (i < size)
{
__m128i s0 = SaveLoadTail(src0 + i, size - i);
__m128i i0 = UnpackU8<0>(s0);
__m128i i1 = UnpackU8<1>(s0);
Update<update, true>(dst + 0 * F, _mm_madd_epi16(alpha, UnpackU8<0>(i0)));
Update<update, true>(dst + 1 * F, _mm_madd_epi16(alpha, UnpackU8<1>(i0)));
Update<update, true>(dst + 2 * F, _mm_madd_epi16(alpha, UnpackU8<0>(i1)));
Update<update, true>(dst + 3 * F, _mm_madd_epi16(alpha, UnpackU8<1>(i1)));
}
}
template<UpdateType update> SIMD_INLINE void ResizerByteArea1x1RowUpdate(const uint8_t * src0, size_t stride, size_t size, int32_t a0, int32_t a1, int32_t * dst)
{
__m128i alpha = SetInt16(a0, a1);
const uint8_t * src1 = src0 + stride;
size_t sizeA = AlignLo(size, A);
size_t i = 0;
for (; i < sizeA; i += A, dst += A)
{
__m128i s0 = _mm_loadu_si128((__m128i*)(src0 + i));
__m128i s1 = _mm_loadu_si128((__m128i*)(src1 + i));
__m128i i0 = UnpackU8<0>(s0, s1);
__m128i i1 = UnpackU8<1>(s0, s1);
Update<update, true>(dst + 0 * F, _mm_madd_epi16(alpha, UnpackU8<0>(i0)));
Update<update, true>(dst + 1 * F, _mm_madd_epi16(alpha, UnpackU8<1>(i0)));
Update<update, true>(dst + 2 * F, _mm_madd_epi16(alpha, UnpackU8<0>(i1)));
Update<update, true>(dst + 3 * F, _mm_madd_epi16(alpha, UnpackU8<1>(i1)));
}
if (i < size)
{
__m128i s0 = _mm_loadu_si128((__m128i*)(src0 + i));
__m128i s1 = SaveLoadTail(src1 + i, size - i);
__m128i i0 = UnpackU8<0>(s0, s1);
__m128i i1 = UnpackU8<1>(s0, s1);
Update<update, true>(dst + 0 * F, _mm_madd_epi16(alpha, UnpackU8<0>(i0)));
Update<update, true>(dst + 1 * F, _mm_madd_epi16(alpha, UnpackU8<1>(i0)));
Update<update, true>(dst + 2 * F, _mm_madd_epi16(alpha, UnpackU8<0>(i1)));
Update<update, true>(dst + 3 * F, _mm_madd_epi16(alpha, UnpackU8<1>(i1)));
}
}
SIMD_INLINE void ResizerByteArea1x1RowSum(const uint8_t * src, size_t stride, size_t count, size_t size, int32_t curr, int32_t zero, int32_t next, int32_t * dst)
{
if (count)
{
size_t i = 0;
ResizerByteArea1x1RowUpdate<UpdateSet>(src, stride, size, curr, count == 1 ? zero - next : zero, dst), src += 2 * stride, i +=2;
for (; i < count; i += 2, src += 2 * stride)
ResizerByteArea1x1RowUpdate<UpdateAdd>(src, stride, size, zero, i == count - 1 ? zero - next : zero, dst);
if (i == count)
ResizerByteArea1x1RowUpdate<UpdateAdd>(src, size, zero - next, dst);
}
else
ResizerByteArea1x1RowUpdate<UpdateSet>(src, size, curr - next, dst);
}
template<size_t N> void ResizerByteArea1x1::Run(const uint8_t * src, size_t srcStride, uint8_t * dst, size_t dstStride)
{
size_t bodyW = _param.dstW - (N == 3 ? 1 : 0), rowSize = _param.srcW * N, rowRest = dstStride - _param.dstW * N;
const int32_t * iy = _iy.data, * ix = _ix.data, * ay = _ay.data, * ax = _ax.data;
int32_t ay0 = ay[0], ax0 = ax[0];
for (size_t dy = 0; dy < _param.dstH; dy++, dst += rowRest)
{
int32_t * buf = _by.data;
size_t yn = iy[dy + 1] - iy[dy];
ResizerByteArea1x1RowSum(src, srcStride, yn, rowSize, ay[dy], ay0, ay[dy + 1], buf), src += yn * srcStride;
size_t dx = 0;
for (; dx < bodyW; dx++, dst += N)
{
size_t xn = ix[dx + 1] - ix[dx];
Sse41::ResizerByteAreaResult<N>(buf, xn, ax[dx], ax0, ax[dx + 1], dst), buf += xn * N;
}
for (; dx < _param.dstW; dx++, dst += N)
{
size_t xn = ix[dx + 1] - ix[dx];
Base::ResizerByteAreaResult<N>(buf, xn, ax[dx], ax0, ax[dx + 1], dst), buf += xn * N;
}
}
}
void ResizerByteArea1x1::Run(const uint8_t * src, size_t srcStride, uint8_t * dst, size_t dstStride)
{
switch (_param.channels)
{
case 1: Run<1>(src, srcStride, dst, dstStride); return;
case 2: Run<2>(src, srcStride, dst, dstStride); return;
case 3: Run<3>(src, srcStride, dst, dstStride); return;
case 4: Run<4>(src, srcStride, dst, dstStride); return;
default:
assert(0);
}
}
//---------------------------------------------------------------------------------------------
ResizerByteArea2x2::ResizerByteArea2x2(const ResParam& param)
: Base::ResizerByteArea2x2(param)
{
}
template<size_t N> SIMD_INLINE __m128i ShuffleColor(__m128i val)
{
return val;
}
template<> SIMD_INLINE __m128i ShuffleColor<2>(__m128i val)
{
static const __m128i IDX = SIMD_MM_SETR_EPI8(0x0, 0x2, 0x1, 0x3, 0x4, 0x6, 0x5, 0x7, 0x8, 0xA, 0x9, 0xB, 0xC, 0xE, 0xD, 0xF);
return _mm_shuffle_epi8(val, IDX);
}
template<> SIMD_INLINE __m128i ShuffleColor<4>(__m128i val)
{
static const __m128i IDX = SIMD_MM_SETR_EPI8(0x0, 0x4, 0x1, 0x5, 0x2, 0x6, 0x3, 0x7, 0x8, 0xC, 0x9, 0xD, 0xA, 0xE, 0xB, 0xF);
return _mm_shuffle_epi8(val, IDX);
}
template<size_t N> SIMD_INLINE __m128i SaveLoadTail2x2(const uint8_t* ptr, size_t tail)
{
uint8_t buffer[DA];
_mm_storeu_si128((__m128i*)(buffer), LoadAfterLast<N>(_mm_loadu_si128((__m128i*)(ptr + tail - A))));
return _mm_loadu_si128((__m128i*)(buffer + A - tail - N));
}
template<size_t N, UpdateType update> SIMD_INLINE void ResizerByteArea2x2RowUpdateColor(const uint8_t* src0, const uint8_t* src1, size_t size, int32_t val, int32_t* dst)
{
if (update == UpdateAdd && val == 0)
return;
size_t size4F = AlignLoAny(size, 4 * F);
__m128i _val = _mm_set1_epi16(val);
size_t i = 0;
for (; i < size4F; i += 4 * F, dst += 2 * F)
{
__m128i s0 = _mm_maddubs_epi16(ShuffleColor<N>(_mm_loadu_si128((__m128i*)(src0 + i))), K8_01);
__m128i s1 = _mm_maddubs_epi16(ShuffleColor<N>(_mm_loadu_si128((__m128i*)(src1 + i))), K8_01);
Update<update, false>(dst + 0, _mm_madd_epi16(_mm_unpacklo_epi16(s0, s1), _val));
Update<update, false>(dst + F, _mm_madd_epi16(_mm_unpackhi_epi16(s0, s1), _val));
}
if (i < size)
{
size_t tail = size - i;
__m128i s0 = _mm_maddubs_epi16(ShuffleColor<N>(SaveLoadTail2x2<N>(src0 + i, tail)), K8_01);
__m128i s1 = _mm_maddubs_epi16(ShuffleColor<N>(SaveLoadTail2x2<N>(src1 + i, tail)), K8_01);
Update<update, false>(dst + 0, _mm_madd_epi16(_mm_unpacklo_epi16(s0, s1), _val));
Update<update, false>(dst + F, _mm_madd_epi16(_mm_unpackhi_epi16(s0, s1), _val));
}
}
template<size_t N> SIMD_INLINE void ResizerByteArea2x2RowSum(const uint8_t* src, size_t stride, size_t count, size_t size, int32_t curr, int32_t zero, int32_t next, bool tail, int32_t* dst)
{
size_t c = 0;
if (count)
{
ResizerByteArea2x2RowUpdateColor<N, UpdateSet>(src, src + stride, size, curr, dst), src += 2 * stride, c += 2;
for (; c < count; c += 2, src += 2 * stride)
ResizerByteArea2x2RowUpdateColor<N, UpdateAdd>(src, src + stride, size, zero, dst);
ResizerByteArea2x2RowUpdateColor<N, UpdateAdd>(src, tail ? src : src + stride, size, zero - next, dst);
}
else
ResizerByteArea2x2RowUpdateColor<N, UpdateSet>(src, tail ? src : src + stride, size, curr - next, dst);
}
SIMD_INLINE void SaveLoadTailBgr2x2(const uint8_t* ptr, size_t tail, __m128i * val)
{
uint8_t buffer[3 * A] = { 0 };
_mm_storeu_si128((__m128i*)(buffer + 0), _mm_loadu_si128((__m128i*)(ptr + tail - 24)));
_mm_storeu_si128((__m128i*)(buffer + 11), LoadAfterLast<3>(_mm_loadu_si128((__m128i*)(ptr + tail - 16))));
val[0] = _mm_loadu_si128((__m128i*)(buffer + 24 - tail));
val[1] = _mm_loadu_si128((__m128i*)(buffer + 32 - tail));
}
template<UpdateType update> SIMD_INLINE void ResizerByteArea2x2RowUpdateBgr(const uint8_t* src0, const uint8_t* src1, size_t size, int32_t val, int32_t* dst)
{
if (update == UpdateAdd && val == 0)
return;
size_t i = 0;
__m128i _val = _mm_set1_epi32(val);
static const __m128i K8_BGR0 = SIMD_MM_SETR_EPI8(0x0, 0x3, 0x1, 0x4, 0x2, 0x5, 0x6, 0x9, -1, -1, -1, -1, 0x7, 0xA, 0x8, 0xB);
static const __m128i K8_BGR1 = SIMD_MM_SETR_EPI8(0x4, 0x7, 0x5, 0x8, -1, -1, -1, -1, 0x6, 0x9, 0xA, 0xD, 0xB, 0xE, 0xC, 0xF);
size_t size24 = AlignLoAny(size, 24);
for (; i < size24; i += 24, dst += 12)
{
__m128i s00 = _mm_maddubs_epi16(_mm_shuffle_epi8(Load<false>((__m128i*)(src0 + i + 0)), K8_BGR0), K8_01);
__m128i s01 = _mm_maddubs_epi16(_mm_shuffle_epi8(Load<false>((__m128i*)(src0 + i + 8)), K8_BGR1), K8_01);
__m128i s10 = _mm_maddubs_epi16(_mm_shuffle_epi8(Load<false>((__m128i*)(src1 + i + 0)), K8_BGR0), K8_01);
__m128i s11 = _mm_maddubs_epi16(_mm_shuffle_epi8(Load<false>((__m128i*)(src1 + i + 8)), K8_BGR1), K8_01);
__m128i s0 = _mm_add_epi16(s00, s10);
__m128i s1 = _mm_add_epi16(s01, s11);
Update<update, false>(dst + 0 * F, _mm_madd_epi16(_mm_cvtepi16_epi32(s0), _val));
Update<update, false>(dst + 1 * F, _mm_madd_epi16(_mm_cvtepi16_epi32(_mm_alignr_epi8(s1, s0, 12)), _val));
Update<update, false>(dst + 2 * F, _mm_madd_epi16(_mm_cvtepi16_epi32(_mm_srli_si128(s1, 8)), _val));
}
if (i < size)
{
size_t tail = size - i;
__m128i s[4];
SaveLoadTailBgr2x2(src0 + i, tail, s + 0);
SaveLoadTailBgr2x2(src1 + i, tail, s + 2);
__m128i s0 = _mm_add_epi16(
_mm_maddubs_epi16(_mm_shuffle_epi8(s[0], K8_BGR0), K8_01),
_mm_maddubs_epi16(_mm_shuffle_epi8(s[2], K8_BGR0), K8_01));
__m128i s1 = _mm_add_epi16(
_mm_maddubs_epi16(_mm_shuffle_epi8(s[1], K8_BGR1), K8_01),
_mm_maddubs_epi16(_mm_shuffle_epi8(s[3], K8_BGR1), K8_01));
Update<update, false>(dst + 0 * F, _mm_madd_epi16(_mm_cvtepi16_epi32(s0), _val));
Update<update, false>(dst + 1 * F, _mm_madd_epi16(_mm_cvtepi16_epi32(_mm_alignr_epi8(s1, s0, 12)), _val));
Update<update, false>(dst + 2 * F, _mm_madd_epi16(_mm_cvtepi16_epi32(_mm_srli_si128(s1, 8)), _val));
}
}
template<> SIMD_INLINE void ResizerByteArea2x2RowSum<3>(const uint8_t* src, size_t stride, size_t count, size_t size, int32_t curr, int32_t zero, int32_t next, bool tail, int32_t* dst)
{
size_t c = 0;
if (count)
{
ResizerByteArea2x2RowUpdateBgr<UpdateSet>(src, src + stride, size, curr, dst), src += 2 * stride, c += 2;
for (; c < count; c += 2, src += 2 * stride)
ResizerByteArea2x2RowUpdateBgr<UpdateAdd>(src, src + stride, size, zero, dst);
ResizerByteArea2x2RowUpdateBgr<UpdateAdd>(src, tail ? src : src + stride, size, zero - next, dst);
}
else
ResizerByteArea2x2RowUpdateBgr<UpdateSet>(src, tail ? src : src + stride, size, curr - next, dst);
}
template<size_t N> void ResizerByteArea2x2::Run(const uint8_t* src, size_t srcStride, uint8_t* dst, size_t dstStride)
{
size_t bodyW = _param.dstW - (N == 3 ? 1 : 0), rowSize = _param.srcW * N, rowRest = dstStride - _param.dstW * N;
const int32_t* iy = _iy.data, * ix = _ix.data, * ay = _ay.data, * ax = _ax.data;
int32_t ay0 = ay[0], ax0 = ax[0];
for (size_t dy = 0; dy < _param.dstH; dy++, dst += rowRest)
{
int32_t* buf = _by.data;
size_t yn = (iy[dy + 1] - iy[dy]) * 2;
bool tail = (dy == _param.dstH - 1) && (_param.srcH & 1);
ResizerByteArea2x2RowSum<N>(src, srcStride, yn, rowSize, ay[dy], ay0, ay[dy + 1], tail, buf), src += yn * srcStride;
size_t dx = 0;
for (; dx < bodyW; dx++, dst += N)
{
size_t xn = ix[dx + 1] - ix[dx];
Sse41::ResizerByteAreaResult<N>(buf, xn, ax[dx], ax0, ax[dx + 1], dst), buf += xn * N;
}
for (; dx < _param.dstW; dx++, dst += N)
{
size_t xn = ix[dx + 1] - ix[dx];
Base::ResizerByteAreaResult<N>(buf, xn, ax[dx], ax0, ax[dx + 1], dst), buf += xn * N;
}
}
}
void ResizerByteArea2x2::Run(const uint8_t* src, size_t srcStride, uint8_t* dst, size_t dstStride)
{
switch (_param.channels)
{
case 1: Run<1>(src, srcStride, dst, dstStride); return;
case 2: Run<2>(src, srcStride, dst, dstStride); return;
case 3: Run<3>(src, srcStride, dst, dstStride); return;
case 4: Run<4>(src, srcStride, dst, dstStride); return;
default:
assert(0);
}
}
}
#endif
}
|
