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
|
/*
* Copyright (c) 2010 The WebM project authors. All Rights Reserved.
*
* Use of this source code is governed by a BSD-style license
* that can be found in the LICENSE file in the root of the source
* tree. An additional intellectual property rights grant can be found
* in the file PATENTS. All contributing project authors may
* be found in the AUTHORS file in the root of the source tree.
*/
#include "./vpx_dsp_rtcd.h"
#include "vpx_ports/mem.h"
DECLARE_ALIGNED(16, static const int16_t, bilinear_filters_mmx[8][8]) = {
{ 128, 128, 128, 128, 0, 0, 0, 0 },
{ 112, 112, 112, 112, 16, 16, 16, 16 },
{ 96, 96, 96, 96, 32, 32, 32, 32 },
{ 80, 80, 80, 80, 48, 48, 48, 48 },
{ 64, 64, 64, 64, 64, 64, 64, 64 },
{ 48, 48, 48, 48, 80, 80, 80, 80 },
{ 32, 32, 32, 32, 96, 96, 96, 96 },
{ 16, 16, 16, 16, 112, 112, 112, 112 }
};
extern void vpx_get4x4var_mmx(const uint8_t *a, int a_stride,
const uint8_t *b, int b_stride,
unsigned int *sse, int *sum);
extern void vpx_filter_block2d_bil4x4_var_mmx(const unsigned char *ref_ptr,
int ref_pixels_per_line,
const unsigned char *src_ptr,
int src_pixels_per_line,
const int16_t *HFilter,
const int16_t *VFilter,
int *sum,
unsigned int *sumsquared);
extern void vpx_filter_block2d_bil_var_mmx(const unsigned char *ref_ptr,
int ref_pixels_per_line,
const unsigned char *src_ptr,
int src_pixels_per_line,
unsigned int Height,
const int16_t *HFilter,
const int16_t *VFilter,
int *sum,
unsigned int *sumsquared);
unsigned int vpx_variance4x4_mmx(const unsigned char *a, int a_stride,
const unsigned char *b, int b_stride,
unsigned int *sse) {
unsigned int var;
int avg;
vpx_get4x4var_mmx(a, a_stride, b, b_stride, &var, &avg);
*sse = var;
return (var - (((unsigned int)avg * avg) >> 4));
}
unsigned int vpx_variance8x8_mmx(const unsigned char *a, int a_stride,
const unsigned char *b, int b_stride,
unsigned int *sse) {
unsigned int var;
int avg;
vpx_get8x8var_mmx(a, a_stride, b, b_stride, &var, &avg);
*sse = var;
return (var - (((unsigned int)avg * avg) >> 6));
}
unsigned int vpx_mse16x16_mmx(const unsigned char *a, int a_stride,
const unsigned char *b, int b_stride,
unsigned int *sse) {
unsigned int sse0, sse1, sse2, sse3, var;
int sum0, sum1, sum2, sum3;
vpx_get8x8var_mmx(a, a_stride, b, b_stride, &sse0, &sum0);
vpx_get8x8var_mmx(a + 8, a_stride, b + 8, b_stride, &sse1, &sum1);
vpx_get8x8var_mmx(a + 8 * a_stride, a_stride,
b + 8 * b_stride, b_stride, &sse2, &sum2);
vpx_get8x8var_mmx(a + 8 * a_stride + 8, a_stride,
b + 8 * b_stride + 8, b_stride, &sse3, &sum3);
var = sse0 + sse1 + sse2 + sse3;
*sse = var;
return var;
}
unsigned int vpx_variance16x16_mmx(const unsigned char *a, int a_stride,
const unsigned char *b, int b_stride,
unsigned int *sse) {
unsigned int sse0, sse1, sse2, sse3, var;
int sum0, sum1, sum2, sum3, avg;
vpx_get8x8var_mmx(a, a_stride, b, b_stride, &sse0, &sum0);
vpx_get8x8var_mmx(a + 8, a_stride, b + 8, b_stride, &sse1, &sum1);
vpx_get8x8var_mmx(a + 8 * a_stride, a_stride,
b + 8 * b_stride, b_stride, &sse2, &sum2);
vpx_get8x8var_mmx(a + 8 * a_stride + 8, a_stride,
b + 8 * b_stride + 8, b_stride, &sse3, &sum3);
var = sse0 + sse1 + sse2 + sse3;
avg = sum0 + sum1 + sum2 + sum3;
*sse = var;
return (var - (((unsigned int)avg * avg) >> 8));
}
unsigned int vpx_variance16x8_mmx(const unsigned char *a, int a_stride,
const unsigned char *b, int b_stride,
unsigned int *sse) {
unsigned int sse0, sse1, var;
int sum0, sum1, avg;
vpx_get8x8var_mmx(a, a_stride, b, b_stride, &sse0, &sum0);
vpx_get8x8var_mmx(a + 8, a_stride, b + 8, b_stride, &sse1, &sum1);
var = sse0 + sse1;
avg = sum0 + sum1;
*sse = var;
return (var - (((unsigned int)avg * avg) >> 7));
}
unsigned int vpx_variance8x16_mmx(const unsigned char *a, int a_stride,
const unsigned char *b, int b_stride,
unsigned int *sse) {
unsigned int sse0, sse1, var;
int sum0, sum1, avg;
vpx_get8x8var_mmx(a, a_stride, b, b_stride, &sse0, &sum0);
vpx_get8x8var_mmx(a + 8 * a_stride, a_stride,
b + 8 * b_stride, b_stride, &sse1, &sum1);
var = sse0 + sse1;
avg = sum0 + sum1;
*sse = var;
return (var - (((unsigned int)avg * avg) >> 7));
}
uint32_t vpx_sub_pixel_variance4x4_mmx(const uint8_t *a, int a_stride,
int xoffset, int yoffset,
const uint8_t *b, int b_stride,
uint32_t *sse) {
int xsum;
unsigned int xxsum;
vpx_filter_block2d_bil4x4_var_mmx(a, a_stride, b, b_stride,
bilinear_filters_mmx[xoffset],
bilinear_filters_mmx[yoffset],
&xsum, &xxsum);
*sse = xxsum;
return (xxsum - (((unsigned int)xsum * xsum) >> 4));
}
uint32_t vpx_sub_pixel_variance8x8_mmx(const uint8_t *a, int a_stride,
int xoffset, int yoffset,
const uint8_t *b, int b_stride,
uint32_t *sse) {
int xsum;
uint32_t xxsum;
vpx_filter_block2d_bil_var_mmx(a, a_stride, b, b_stride, 8,
bilinear_filters_mmx[xoffset],
bilinear_filters_mmx[yoffset],
&xsum, &xxsum);
*sse = xxsum;
return (xxsum - (((uint32_t)xsum * xsum) >> 6));
}
uint32_t vpx_sub_pixel_variance16x16_mmx(const uint8_t *a, int a_stride,
int xoffset, int yoffset,
const uint8_t *b, int b_stride,
uint32_t *sse) {
int xsum0, xsum1;
unsigned int xxsum0, xxsum1;
vpx_filter_block2d_bil_var_mmx(a, a_stride, b, b_stride, 16,
bilinear_filters_mmx[xoffset],
bilinear_filters_mmx[yoffset],
&xsum0, &xxsum0);
vpx_filter_block2d_bil_var_mmx(a + 8, a_stride, b + 8, b_stride, 16,
bilinear_filters_mmx[xoffset],
bilinear_filters_mmx[yoffset],
&xsum1, &xxsum1);
xsum0 += xsum1;
xxsum0 += xxsum1;
*sse = xxsum0;
return (xxsum0 - (((uint32_t)xsum0 * xsum0) >> 8));
}
uint32_t vpx_sub_pixel_variance16x8_mmx(const uint8_t *a, int a_stride,
int xoffset, int yoffset,
const uint8_t *b, int b_stride,
uint32_t *sse) {
int xsum0, xsum1;
unsigned int xxsum0, xxsum1;
vpx_filter_block2d_bil_var_mmx(a, a_stride, b, b_stride, 8,
bilinear_filters_mmx[xoffset],
bilinear_filters_mmx[yoffset],
&xsum0, &xxsum0);
vpx_filter_block2d_bil_var_mmx(a + 8, a_stride, b + 8, b_stride, 8,
bilinear_filters_mmx[xoffset],
bilinear_filters_mmx[yoffset],
&xsum1, &xxsum1);
xsum0 += xsum1;
xxsum0 += xxsum1;
*sse = xxsum0;
return (xxsum0 - (((uint32_t)xsum0 * xsum0) >> 7));
}
uint32_t vpx_sub_pixel_variance8x16_mmx(const uint8_t *a, int a_stride,
int xoffset, int yoffset,
const uint8_t *b, int b_stride,
uint32_t *sse) {
int xsum;
unsigned int xxsum;
vpx_filter_block2d_bil_var_mmx(a, a_stride, b, b_stride, 16,
bilinear_filters_mmx[xoffset],
bilinear_filters_mmx[yoffset],
&xsum, &xxsum);
*sse = xxsum;
return (xxsum - (((uint32_t)xsum * xsum) >> 7));
}
uint32_t vpx_variance_halfpixvar16x16_h_mmx(const uint8_t *a, int a_stride,
const uint8_t *b, int b_stride,
uint32_t *sse) {
return vpx_sub_pixel_variance16x16_mmx(a, a_stride, 4, 0, b, b_stride, sse);
}
uint32_t vpx_variance_halfpixvar16x16_v_mmx(const uint8_t *a, int a_stride,
const uint8_t *b, int b_stride,
uint32_t *sse) {
return vpx_sub_pixel_variance16x16_mmx(a, a_stride, 0, 4, b, b_stride, sse);
}
uint32_t vpx_variance_halfpixvar16x16_hv_mmx(const uint8_t *a, int a_stride,
const uint8_t *b, int b_stride,
uint32_t *sse) {
return vpx_sub_pixel_variance16x16_mmx(a, a_stride, 4, 4, b, b_stride, sse);
}
|
