FazBrowse GitHub Viewer
|
Trending
|
URL:
|
Home
Tools:
[Download Repo ZIP]
[View Raw Code]
[Original HTTPS Page]
opencv/modules/core/src/convert.sse4_1.cpp at master · pythonmjs/opencv · GitHub
pythonmjs
/
opencv
Public
forked from
opencv/opencv
Notifications
You must be signed in to change notification settings
Fork
0
Star
0
Code
Pull requests
0
Actions
Projects
Wiki
Security and quality
0
Insights
Additional navigation options
Code
Pull requests
Actions
Projects
Wiki
Security and quality
Insights
Expand file tree
Breadcrumbs
opencv
/
modules
/
core
/
src
/
convert.sse4_1.cpp
Copy path
More file actions
More file actions
Latest commit
History
History
History
203 lines (151 loc) · 6.83 KB
Breadcrumbs
opencv
/
modules
/
core
/
src
/
convert.sse4_1.cpp
Copy path
File metadata and controls
203 lines (151 loc) · 6.83 KB
Raw
Copy raw file
Download raw file
Open symbols panel
Edit and raw actions
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
//
This file is part of OpenCV project.
//
It is subject to the license terms in the LICENSE file found in the top-level directory
//
of this distribution and at http://opencv.org/license.html
#
include
"
precomp.hpp
"
#
include
"
convert.hpp
"
namespace
cv
{
namespace
opt_SSE4_1
{
int
cvtScale_SIMD_u8u16f32_SSE41
(
const
uchar * src, ushort * dst,
int
width,
float
scale,
float
shift)
{
int
x =
0
;
__m128i v_zero =
_mm_setzero_si128
();
__m128 v_scale =
_mm_set1_ps
(scale), v_shift =
_mm_set1_ps
(shift);
for
( ; x <= width -
8
; x +=
8
)
{
__m128i v_src =
_mm_unpacklo_epi8
(
_mm_loadl_epi64
((__m128i
const
*)(src + x)), v_zero);
__m128 v_src_f =
_mm_cvtepi32_ps
(
_mm_unpacklo_epi16
(v_src, v_zero));
__m128 v_dst_0 =
_mm_add_ps
(
_mm_mul_ps
(v_src_f, v_scale), v_shift);
v_src_f =
_mm_cvtepi32_ps
(
_mm_unpackhi_epi16
(v_src, v_zero));
__m128 v_dst_1 =
_mm_add_ps
(
_mm_mul_ps
(v_src_f, v_scale), v_shift);
__m128i v_dst =
_mm_packus_epi32
(
_mm_cvtps_epi32
(v_dst_0),
_mm_cvtps_epi32
(v_dst_1));
_mm_storeu_si128
((__m128i *)(dst + x), v_dst);
}
return
x;
}
int
cvtScale_SIMD_s8u16f32_SSE41
(
const
schar * src, ushort * dst,
int
width,
float
scale,
float
shift)
{
int
x =
0
;
__m128i v_zero =
_mm_setzero_si128
();
__m128 v_scale =
_mm_set1_ps
(scale), v_shift =
_mm_set1_ps
(shift);
for
( ; x <= width -
8
; x +=
8
)
{
__m128i v_src =
_mm_srai_epi16
(
_mm_unpacklo_epi8
(v_zero,
_mm_loadl_epi64
((__m128i
const
*)(src + x))),
8
);
__m128 v_src_f =
_mm_cvtepi32_ps
(
_mm_srai_epi32
(
_mm_unpacklo_epi16
(v_zero, v_src),
16
));
__m128 v_dst_0 =
_mm_add_ps
(
_mm_mul_ps
(v_src_f, v_scale), v_shift);
v_src_f =
_mm_cvtepi32_ps
(
_mm_srai_epi32
(
_mm_unpackhi_epi16
(v_zero, v_src),
16
));
__m128 v_dst_1 =
_mm_add_ps
(
_mm_mul_ps
(v_src_f, v_scale), v_shift);
__m128i v_dst =
_mm_packus_epi32
(
_mm_cvtps_epi32
(v_dst_0),
_mm_cvtps_epi32
(v_dst_1));
_mm_storeu_si128
((__m128i *)(dst + x), v_dst);
}
return
x;
}
int
cvtScale_SIMD_u16u16f32_SSE41
(
const
ushort * src, ushort * dst,
int
width,
float
scale,
float
shift)
{
int
x =
0
;
__m128i v_zero =
_mm_setzero_si128
();
__m128 v_scale =
_mm_set1_ps
(scale), v_shift =
_mm_set1_ps
(shift);
for
( ; x <= width -
8
; x +=
8
)
{
__m128i v_src =
_mm_loadu_si128
((__m128i
const
*)(src + x));
__m128 v_src_f =
_mm_cvtepi32_ps
(
_mm_unpacklo_epi16
(v_src, v_zero));
__m128 v_dst_0 =
_mm_add_ps
(
_mm_mul_ps
(v_src_f, v_scale), v_shift);
v_src_f =
_mm_cvtepi32_ps
(
_mm_unpackhi_epi16
(v_src, v_zero));
__m128 v_dst_1 =
_mm_add_ps
(
_mm_mul_ps
(v_src_f, v_scale), v_shift);
__m128i v_dst =
_mm_packus_epi32
(
_mm_cvtps_epi32
(v_dst_0),
_mm_cvtps_epi32
(v_dst_1));
_mm_storeu_si128
((__m128i *)(dst + x), v_dst);
}
return
x;
}
int
cvtScale_SIMD_s16u16f32_SSE41
(
const
short
* src, ushort * dst,
int
width,
float
scale,
float
shift)
{
int
x =
0
;
__m128i v_zero =
_mm_setzero_si128
();
__m128 v_scale =
_mm_set1_ps
(scale), v_shift =
_mm_set1_ps
(shift);
for
( ; x <= width -
8
; x +=
8
)
{
__m128i v_src =
_mm_loadu_si128
((__m128i
const
*)(src + x));
__m128 v_src_f =
_mm_cvtepi32_ps
(
_mm_srai_epi32
(
_mm_unpacklo_epi16
(v_zero, v_src),
16
));
__m128 v_dst_0 =
_mm_add_ps
(
_mm_mul_ps
(v_src_f, v_scale), v_shift);
v_src_f =
_mm_cvtepi32_ps
(
_mm_srai_epi32
(
_mm_unpackhi_epi16
(v_zero, v_src),
16
));
__m128 v_dst_1 =
_mm_add_ps
(
_mm_mul_ps
(v_src_f, v_scale), v_shift);
__m128i v_dst =
_mm_packus_epi32
(
_mm_cvtps_epi32
(v_dst_0),
_mm_cvtps_epi32
(v_dst_1));
_mm_storeu_si128
((__m128i *)(dst + x), v_dst);
}
return
x;
}
int
cvtScale_SIMD_s32u16f32_SSE41
(
const
int
* src, ushort * dst,
int
width,
float
scale,
float
shift)
{
int
x =
0
;
__m128 v_scale =
_mm_set1_ps
(scale), v_shift =
_mm_set1_ps
(shift);
for
( ; x <= width -
8
; x +=
8
)
{
__m128i v_src =
_mm_loadu_si128
((__m128i
const
*)(src + x));
__m128 v_dst_0 =
_mm_add_ps
(
_mm_mul_ps
(
_mm_cvtepi32_ps
(v_src), v_scale), v_shift);
v_src =
_mm_loadu_si128
((__m128i
const
*)(src + x +
4
));
__m128 v_dst_1 =
_mm_add_ps
(
_mm_mul_ps
(
_mm_cvtepi32_ps
(v_src), v_scale), v_shift);
__m128i v_dst =
_mm_packus_epi32
(
_mm_cvtps_epi32
(v_dst_0),
_mm_cvtps_epi32
(v_dst_1));
_mm_storeu_si128
((__m128i *)(dst + x), v_dst);
}
return
x;
}
int
cvtScale_SIMD_f32u16f32_SSE41
(
const
float
* src, ushort * dst,
int
width,
float
scale,
float
shift)
{
int
x =
0
;
__m128 v_scale =
_mm_set1_ps
(scale), v_shift =
_mm_set1_ps
(shift);
for
( ; x <= width -
8
; x +=
8
)
{
__m128 v_src =
_mm_loadu_ps
(src + x);
__m128 v_dst_0 =
_mm_add_ps
(
_mm_mul_ps
(v_src, v_scale), v_shift);
v_src =
_mm_loadu_ps
(src + x +
4
);
__m128 v_dst_1 =
_mm_add_ps
(
_mm_mul_ps
(v_src, v_scale), v_shift);
__m128i v_dst =
_mm_packus_epi32
(
_mm_cvtps_epi32
(v_dst_0),
_mm_cvtps_epi32
(v_dst_1));
_mm_storeu_si128
((__m128i *)(dst + x), v_dst);
}
return
x;
}
int
cvtScale_SIMD_f64u16f32_SSE41
(
const
double
* src, ushort * dst,
int
width,
float
scale,
float
shift)
{
int
x =
0
;
__m128 v_scale =
_mm_set1_ps
(scale), v_shift =
_mm_set1_ps
(shift);
for
( ; x <= width -
8
; x +=
8
)
{
__m128 v_src =
_mm_movelh_ps
(
_mm_cvtpd_ps
(
_mm_loadu_pd
(src + x)),
_mm_cvtpd_ps
(
_mm_loadu_pd
(src + x +
2
)));
__m128 v_dst_0 =
_mm_add_ps
(
_mm_mul_ps
(v_src, v_scale), v_shift);
v_src =
_mm_movelh_ps
(
_mm_cvtpd_ps
(
_mm_loadu_pd
(src + x +
4
)),
_mm_cvtpd_ps
(
_mm_loadu_pd
(src + x +
6
)));
__m128 v_dst_1 =
_mm_add_ps
(
_mm_mul_ps
(v_src, v_scale), v_shift);
__m128i v_dst =
_mm_packus_epi32
(
_mm_cvtps_epi32
(v_dst_0),
_mm_cvtps_epi32
(v_dst_1));
_mm_storeu_si128
((__m128i *)(dst + x), v_dst);
}
return
x;
}
int
Cvt_SIMD_f64u16_SSE41
(
const
double
* src, ushort * dst,
int
width)
{
int
x =
0
;
for
( ; x <= width -
8
; x +=
8
)
{
__m128 v_src0 =
_mm_cvtpd_ps
(
_mm_loadu_pd
(src + x));
__m128 v_src1 =
_mm_cvtpd_ps
(
_mm_loadu_pd
(src + x +
2
));
__m128 v_src2 =
_mm_cvtpd_ps
(
_mm_loadu_pd
(src + x +
4
));
__m128 v_src3 =
_mm_cvtpd_ps
(
_mm_loadu_pd
(src + x +
6
));
v_src0 =
_mm_movelh_ps
(v_src0, v_src1);
v_src1 =
_mm_movelh_ps
(v_src2, v_src3);
__m128i v_dst =
_mm_packus_epi32
(
_mm_cvtps_epi32
(v_src0),
_mm_cvtps_epi32
(v_src1));
_mm_storeu_si128
((__m128i *)(dst + x), v_dst);
}
return
x;
}
}
}
//
cv::
/*
End of file.
*/
Back
|
FazBrowse Home
|
New Git URL