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
|
;*****************************************************************************
;* x86-optimized functions for interlace filter
;*
;* Copyright (C) 2014 Kieran Kunhya <kierank@obe.tv>
;* Copyright (c) 2014 Michael Niedermayer <michaelni@gmx.at>
;* Copyright (c) 2017 Thomas Mundt <tmundt75@gmail.com>
;*
;* This file is part of FFmpeg.
;*
;* FFmpeg is free software; you can redistribute it and/or modify
;* it under the terms of the GNU General Public License as published by
;* the Free Software Foundation; either version 2 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 General Public License for more details.
;*
;* You should have received a copy of the GNU 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
pw_4: times 8 dw 4
SECTION .text
%macro LOWPASS 1
add dstq, hq
add srcq, hq
add mrefq, srcq
add prefq, srcq
neg hq
pcmpeq%1 m6, m6
.loop:
mova m0, [mrefq+hq]
mova m1, [mrefq+hq+mmsize]
pavg%1 m0, [prefq+hq]
pavg%1 m1, [prefq+hq+mmsize]
pxor m0, m6
pxor m1, m6
pxor m2, m6, [srcq+hq]
pxor m3, m6, [srcq+hq+mmsize]
pavg%1 m0, m2
pavg%1 m1, m3
pxor m0, m6
pxor m1, m6
mova [dstq+hq], m0
mova [dstq+hq+mmsize], m1
add hq, 2*mmsize
jl .loop
REP_RET
%endmacro
%macro LOWPASS_LINE 0
cglobal lowpass_line, 5, 5, 7, dst, h, src, mref, pref
LOWPASS b
cglobal lowpass_line_16, 5, 5, 7, dst, h, src, mref, pref
shl hq, 1
LOWPASS w
%endmacro
%macro LOWPASS_LINE_COMPLEX 0
cglobal lowpass_line_complex, 5, 5, 8, dst, h, src, mref, pref
pxor m7, m7
.loop:
mova m0, [srcq+mrefq]
mova m2, [srcq+prefq]
mova m1, m0
mova m3, m2
punpcklbw m0, m7
punpcklbw m2, m7
punpckhbw m1, m7
punpckhbw m3, m7
paddw m0, m2
paddw m1, m3
mova m6, m0
mova m5, m1
mova m2, [srcq]
mova m3, m2
punpcklbw m2, m7
punpckhbw m3, m7
paddw m0, m2
paddw m1, m3
psllw m2, 1
psllw m3, 1
paddw m0, m2
paddw m1, m3
psllw m0, 1
psllw m1, 1
pcmpgtw m6, m2
pcmpgtw m5, m3
packsswb m6, m5
mova m2, [srcq+mrefq*2]
mova m4, [srcq+prefq*2]
mova m3, m2
mova m5, m4
punpcklbw m2, m7
punpcklbw m4, m7
punpckhbw m3, m7
punpckhbw m5, m7
paddw m2, m4
paddw m3, m5
paddw m0, [pw_4]
paddw m1, [pw_4]
psubusw m0, m2
psubusw m1, m3
psrlw m0, 3
psrlw m1, 3
packuswb m0, m1
mova m1, m0
pmaxub m0, [srcq]
pminub m1, [srcq]
pand m0, m6
pandn m6, m1
por m0, m6
mova [dstq], m0
add dstq, mmsize
add srcq, mmsize
sub hd, mmsize
jg .loop
REP_RET
cglobal lowpass_line_complex_12, 5, 5, 8, 16, dst, h, src, mref, pref, clip_max
movd m7, DWORD clip_maxm
SPLATW m7, m7, 0
mova [rsp], m7
.loop:
mova m0, [srcq+mrefq]
mova m1, [srcq+mrefq+mmsize]
mova m2, [srcq+prefq]
mova m3, [srcq+prefq+mmsize]
paddw m0, m2
paddw m1, m3
mova m6, m0
mova m7, m1
mova m2, [srcq]
mova m3, [srcq+mmsize]
paddw m0, m2
paddw m1, m3
psllw m2, 1
psllw m3, 1
paddw m0, m2
paddw m1, m3
psllw m0, 1
psllw m1, 1
pcmpgtw m6, m2
pcmpgtw m7, m3
mova m2, [srcq+2*mrefq]
mova m3, [srcq+2*mrefq+mmsize]
mova m4, [srcq+2*prefq]
mova m5, [srcq+2*prefq+mmsize]
paddw m2, m4
paddw m3, m5
paddw m0, [pw_4]
paddw m1, [pw_4]
psubusw m0, m2
psubusw m1, m3
psrlw m0, 3
psrlw m1, 3
pminsw m0, [rsp]
pminsw m1, [rsp]
mova m2, m0
mova m3, m1
pmaxsw m0, [srcq]
pmaxsw m1, [srcq+mmsize]
pminsw m2, [srcq]
pminsw m3, [srcq+mmsize]
pand m0, m6
pand m1, m7
pandn m6, m2
pandn m7, m3
por m0, m6
por m1, m7
mova [dstq], m0
mova [dstq+mmsize], m1
add dstq, 2*mmsize
add srcq, 2*mmsize
sub hd, mmsize
jg .loop
REP_RET
%endmacro
INIT_XMM sse2
LOWPASS_LINE
INIT_XMM avx
LOWPASS_LINE
INIT_XMM sse2
LOWPASS_LINE_COMPLEX
|
