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
|
/*
* Loongson LSX optimized add_residual functions for HEVC decoding
*
* Copyright (c) 2023 Loongson Technology Corporation Limited
* Contributed by jinbo <jinbo@loongson.cn>
*
* 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 "loongson_asm.S"
/*
* void ff_hevc_add_residual4x4_lsx(uint8_t *dst, const int16_t *res, ptrdiff_t stride)
*/
.macro ADD_RES_LSX_4x4_8
vldrepl.w vr0, a0, 0
add.d t0, a0, a2
vldrepl.w vr1, t0, 0
vld vr2, a1, 0
vilvl.w vr1, vr1, vr0
vsllwil.hu.bu vr1, vr1, 0
vadd.h vr1, vr1, vr2
vssrani.bu.h vr1, vr1, 0
vstelm.w vr1, a0, 0, 0
vstelm.w vr1, t0, 0, 1
.endm
function ff_hevc_add_residual4x4_8_lsx
ADD_RES_LSX_4x4_8
alsl.d a0, a2, a0, 1
addi.d a1, a1, 16
ADD_RES_LSX_4x4_8
endfunc
/*
* void ff_hevc_add_residual8x8_8_lsx(uint8_t *dst, const int16_t *res, ptrdiff_t stride)
*/
.macro ADD_RES_LSX_8x8_8
vldrepl.d vr0, a0, 0
add.d t0, a0, a2
vldrepl.d vr1, t0, 0
add.d t1, t0, a2
vldrepl.d vr2, t1, 0
add.d t2, t1, a2
vldrepl.d vr3, t2, 0
vld vr4, a1, 0
addi.d t3, zero, 16
vldx vr5, a1, t3
addi.d t4, a1, 32
vld vr6, t4, 0
vldx vr7, t4, t3
vsllwil.hu.bu vr0, vr0, 0
vsllwil.hu.bu vr1, vr1, 0
vsllwil.hu.bu vr2, vr2, 0
vsllwil.hu.bu vr3, vr3, 0
vadd.h vr0, vr0, vr4
vadd.h vr1, vr1, vr5
vadd.h vr2, vr2, vr6
vadd.h vr3, vr3, vr7
vssrani.bu.h vr1, vr0, 0
vssrani.bu.h vr3, vr2, 0
vstelm.d vr1, a0, 0, 0
vstelm.d vr1, t0, 0, 1
vstelm.d vr3, t1, 0, 0
vstelm.d vr3, t2, 0, 1
.endm
function ff_hevc_add_residual8x8_8_lsx
ADD_RES_LSX_8x8_8
alsl.d a0, a2, a0, 2
addi.d a1, a1, 64
ADD_RES_LSX_8x8_8
endfunc
/*
* void ff_hevc_add_residual16x16_8_lsx(uint8_t *dst, const int16_t *res, ptrdiff_t stride)
*/
function ff_hevc_add_residual16x16_8_lsx
.rept 8
vld vr0, a0, 0
vldx vr2, a0, a2
vld vr4, a1, 0
addi.d t0, zero, 16
vldx vr5, a1, t0
addi.d t1, a1, 32
vld vr6, t1, 0
vldx vr7, t1, t0
vexth.hu.bu vr1, vr0
vsllwil.hu.bu vr0, vr0, 0
vexth.hu.bu vr3, vr2
vsllwil.hu.bu vr2, vr2, 0
vadd.h vr0, vr0, vr4
vadd.h vr1, vr1, vr5
vadd.h vr2, vr2, vr6
vadd.h vr3, vr3, vr7
vssrani.bu.h vr1, vr0, 0
vssrani.bu.h vr3, vr2, 0
vst vr1, a0, 0
vstx vr3, a0, a2
alsl.d a0, a2, a0, 1
addi.d a1, a1, 64
.endr
endfunc
/*
* void ff_hevc_add_residual32x32_8_lsx(uint8_t *dst, const int16_t *res, ptrdiff_t stride)
*/
function ff_hevc_add_residual32x32_8_lsx
.rept 32
vld vr0, a0, 0
addi.w t0, zero, 16
vldx vr2, a0, t0
vld vr4, a1, 0
vldx vr5, a1, t0
addi.d t1, a1, 32
vld vr6, t1, 0
vldx vr7, t1, t0
vexth.hu.bu vr1, vr0
vsllwil.hu.bu vr0, vr0, 0
vexth.hu.bu vr3, vr2
vsllwil.hu.bu vr2, vr2, 0
vadd.h vr0, vr0, vr4
vadd.h vr1, vr1, vr5
vadd.h vr2, vr2, vr6
vadd.h vr3, vr3, vr7
vssrani.bu.h vr1, vr0, 0
vssrani.bu.h vr3, vr2, 0
vst vr1, a0, 0
vstx vr3, a0, t0
add.d a0, a0, a2
addi.d a1, a1, 64
.endr
endfunc
|
