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
|
;******************************************************************************
;* SIMD lossless video DSP utils
;* Copyright (c) 2008 Loren Merritt
;* Copyright (c) 2014 Michael Niedermayer
;*
;* 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
pb_ef: times 8 db 14,15
pb_67: times 8 db 6, 7
pb_zzzz2323zzzzabab: db -1,-1,-1,-1, 2, 3, 2, 3,-1,-1,-1,-1,10,11,10,11
pb_zzzzzzzz67676767: db -1,-1,-1,-1,-1,-1,-1,-1, 6, 7, 6, 7, 6, 7, 6, 7
SECTION_TEXT
%macro ADD_INT16_LOOP 1 ; %1 = is_aligned
movd m4, maskd
SPLATW m4, m4
add wq, wq
test wq, 2*mmsize - 1
jz %%.tomainloop
%%.wordloop:
sub wq, 2
mov ax, [srcq+wq]
add ax, [dstq+wq]
and ax, maskw
mov [dstq+wq], ax
test wq, 2*mmsize - 1
jnz %%.wordloop
%%.tomainloop:
add srcq, wq
add dstq, wq
neg wq
jz %%.end
%%.loop:
%if %1
mova m0, [srcq+wq]
mova m1, [dstq+wq]
mova m2, [srcq+wq+mmsize]
mova m3, [dstq+wq+mmsize]
%else
movu m0, [srcq+wq]
movu m1, [dstq+wq]
movu m2, [srcq+wq+mmsize]
movu m3, [dstq+wq+mmsize]
%endif
paddw m0, m1
paddw m2, m3
pand m0, m4
pand m2, m4
%if %1
mova [dstq+wq] , m0
mova [dstq+wq+mmsize], m2
%else
movu [dstq+wq] , m0
movu [dstq+wq+mmsize], m2
%endif
add wq, 2*mmsize
jl %%.loop
%%.end:
RET
%endmacro
INIT_MMX mmx
cglobal add_int16, 4,4,5, dst, src, mask, w
ADD_INT16_LOOP 1
INIT_XMM sse2
cglobal add_int16, 4,4,5, dst, src, mask, w
test srcq, mmsize-1
jnz .unaligned
test dstq, mmsize-1
jnz .unaligned
ADD_INT16_LOOP 1
.unaligned:
ADD_INT16_LOOP 0
%macro DIFF_INT16_LOOP 1 ; %1 = is_aligned
movd m4, maskd
SPLATW m4, m4
add wq, wq
test wq, 2*mmsize - 1
jz %%.tomainloop
%%.wordloop:
sub wq, 2
mov ax, [src1q+wq]
sub ax, [src2q+wq]
and ax, maskw
mov [dstq+wq], ax
test wq, 2*mmsize - 1
jnz %%.wordloop
%%.tomainloop:
add src1q, wq
add src2q, wq
add dstq, wq
neg wq
jz %%.end
%%.loop:
%if %1
mova m0, [src1q+wq]
mova m1, [src2q+wq]
mova m2, [src1q+wq+mmsize]
mova m3, [src2q+wq+mmsize]
%else
movu m0, [src1q+wq]
movu m1, [src2q+wq]
movu m2, [src1q+wq+mmsize]
movu m3, [src2q+wq+mmsize]
%endif
psubw m0, m1
psubw m2, m3
pand m0, m4
pand m2, m4
%if %1
mova [dstq+wq] , m0
mova [dstq+wq+mmsize], m2
%else
movu [dstq+wq] , m0
movu [dstq+wq+mmsize], m2
%endif
add wq, 2*mmsize
jl %%.loop
%%.end:
RET
%endmacro
INIT_MMX mmx
cglobal diff_int16, 5,5,5, dst, src1, src2, mask, w
DIFF_INT16_LOOP 1
INIT_XMM sse2
cglobal diff_int16, 5,5,5, dst, src1, src2, mask, w
test src1q, mmsize-1
jnz .unaligned
test src2q, mmsize-1
jnz .unaligned
test dstq, mmsize-1
jnz .unaligned
DIFF_INT16_LOOP 1
.unaligned:
DIFF_INT16_LOOP 0
%macro ADD_HFYU_LEFT_LOOP_INT16 2 ; %1 = dst_is_aligned, %2 = src_is_aligned
add wq, wq
add srcq, wq
add dstq, wq
neg wq
%%.loop:
%if %2
mova m1, [srcq+wq]
%else
movu m1, [srcq+wq]
%endif
mova m2, m1
pslld m1, 16
paddw m1, m2
mova m2, m1
pshufb m1, m3
paddw m1, m2
pshufb m0, m5
%if mmsize == 16
mova m2, m1
pshufb m1, m4
paddw m1, m2
%endif
paddw m0, m1
pand m0, m7
%if %1
mova [dstq+wq], m0
%else
movq [dstq+wq], m0
movhps [dstq+wq+8], m0
%endif
add wq, mmsize
jl %%.loop
mov eax, mmsize-1
sub eax, wd
mov wd, eax
shl wd, 8
lea eax, [wd+eax-1]
movd m1, eax
pshufb m0, m1
movd eax, m0
RET
%endmacro
; int add_hfyu_left_prediction_int16(uint16_t *dst, const uint16_t *src, unsigned mask, int w, int left)
INIT_MMX ssse3
cglobal add_hfyu_left_prediction_int16, 4,4,8, dst, src, mask, w, left
.skip_prologue:
mova m5, [pb_67]
mova m3, [pb_zzzz2323zzzzabab]
movd m0, leftm
psllq m0, 48
movd m7, maskm
SPLATW m7 ,m7
ADD_HFYU_LEFT_LOOP_INT16 1, 1
INIT_XMM sse4
cglobal add_hfyu_left_prediction_int16, 4,4,8, dst, src, mask, w, left
mova m5, [pb_ef]
mova m4, [pb_zzzzzzzz67676767]
mova m3, [pb_zzzz2323zzzzabab]
movd m0, leftm
pslldq m0, 14
movd m7, maskm
SPLATW m7 ,m7
test srcq, 15
jnz .src_unaligned
test dstq, 15
jnz .dst_unaligned
ADD_HFYU_LEFT_LOOP_INT16 1, 1
.dst_unaligned:
ADD_HFYU_LEFT_LOOP_INT16 0, 1
.src_unaligned:
ADD_HFYU_LEFT_LOOP_INT16 0, 0
|
