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
|
;******************************************************************************
;* TAK DSP SIMD optimizations
;*
;* Copyright (C) 2015 Paul B Mahol
;*
;* This file is part of FFmpeg.
;*
;* FFmpeg 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.
;*
;* FFmpeg 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 FFmpeg; if not, write to the Free Software
;* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
;******************************************************************************
%include "libavutil/x86/x86util.asm"
SECTION_RODATA
pd_128: times 4 dd 128
SECTION .text
%macro TAK_DECORRELATE 0
cglobal tak_decorrelate_ls, 3, 3, 2, p1, p2, length
shl lengthd, 2
add p1q, lengthq
add p2q, lengthq
neg lengthq
.loop:
mova m0, [p1q+lengthq+mmsize*0]
mova m1, [p1q+lengthq+mmsize*1]
paddd m0, [p2q+lengthq+mmsize*0]
paddd m1, [p2q+lengthq+mmsize*1]
mova [p2q+lengthq+mmsize*0], m0
mova [p2q+lengthq+mmsize*1], m1
add lengthq, mmsize*2
jl .loop
RET
cglobal tak_decorrelate_sr, 3, 3, 2, p1, p2, length
shl lengthd, 2
add p1q, lengthq
add p2q, lengthq
neg lengthq
.loop:
mova m0, [p2q+lengthq+mmsize*0]
mova m1, [p2q+lengthq+mmsize*1]
psubd m0, [p1q+lengthq+mmsize*0]
psubd m1, [p1q+lengthq+mmsize*1]
mova [p1q+lengthq+mmsize*0], m0
mova [p1q+lengthq+mmsize*1], m1
add lengthq, mmsize*2
jl .loop
RET
cglobal tak_decorrelate_sm, 3, 3, 6, p1, p2, length
shl lengthd, 2
add p1q, lengthq
add p2q, lengthq
neg lengthq
.loop:
mova m0, [p1q+lengthq]
mova m1, [p2q+lengthq]
mova m3, [p1q+lengthq+mmsize]
mova m4, [p2q+lengthq+mmsize]
psrad m2, m1, 1
psrad m5, m4, 1
psubd m0, m2
psubd m3, m5
paddd m1, m0
paddd m4, m3
mova [p1q+lengthq], m0
mova [p2q+lengthq], m1
mova [p1q+lengthq+mmsize], m3
mova [p2q+lengthq+mmsize], m4
add lengthq, mmsize*2
jl .loop
RET
%endmacro
INIT_XMM sse2
TAK_DECORRELATE
%if HAVE_AVX2_EXTERNAL
INIT_YMM avx2
TAK_DECORRELATE
%endif
%macro TAK_DECORRELATE_SF 0
cglobal tak_decorrelate_sf, 3, 3, 5, p1, p2, length, dshift, dfactor
shl lengthd, 2
add p1q, lengthq
add p2q, lengthq
neg lengthq
movd xm2, dshiftm
%if UNIX64
movd xm3, dfactorm
VPBROADCASTD m3, xm3
%else
VPBROADCASTD m3, dfactorm
%endif
VBROADCASTI128 m4, [pd_128]
.loop:
mova m1, [p2q+lengthq]
psrad m1, xm2
pmulld m1, m3
paddd m1, m4
psrad m1, 8
pslld m1, xm2
psubd m1, [p1q+lengthq]
mova [p1q+lengthq], m1
add lengthq, mmsize
jl .loop
RET
%endmacro
INIT_XMM sse4
TAK_DECORRELATE_SF
%if HAVE_AVX2_EXTERNAL
INIT_YMM avx2
TAK_DECORRELATE_SF
%endif
|
