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
|
# this code comes under GPL
# This code was taken from http://www.mpg123.org
# See ChangeLog of mpg123-0.59s-pre.1 for detail
# Applied to mplayer by Nick Kurshev <nickols_k@mail.ru>
#
# Local ChangeLog:
# - Partial loops unrolling and removing MOVW insn from loops
#
.data
.align 8
null_one: .long 0x0000ffff, 0x0000ffff
one_null: .long 0xffff0000, 0xffff0000
.text
.globl synth_1to1_MMX_s
synth_1to1_MMX_s:
pushl %ebp
pushl %edi
pushl %esi
pushl %ebx
movl 24(%esp),%ecx
movl 28(%esp),%edi
movl $15,%ebx
movl 36(%esp),%edx
leal (%edi,%ecx,2),%edi
decl %ecx
movl 32(%esp),%esi
movl (%edx),%eax
jecxz .L1
decl %eax
andl %ebx,%eax
leal 1088(%esi),%esi
movl %eax,(%edx)
.L1:
leal (%esi,%eax,2),%edx
movl %eax,%ebp
incl %eax
pushl 20(%esp)
andl %ebx,%eax
leal 544(%esi,%eax,2),%ecx
incl %ebx
testl $1, %eax
jnz .L2
xchgl %edx,%ecx
incl %ebp
leal 544(%esi),%esi
.L2:
emms
pushl %edx
pushl %ecx
call *dct64_MMX_func
addl $12,%esp
leal 1(%ebx), %ecx
subl %ebp,%ebx
pushl %ecx
leal decwins(%ebx,%ebx,1), %edx
shrl $1, %ecx
.align 16
.L3:
movq (%edx),%mm0
movq 64(%edx),%mm4
pmaddwd (%esi),%mm0
pmaddwd 32(%esi),%mm4
movq 8(%edx),%mm1
movq 72(%edx),%mm5
pmaddwd 8(%esi),%mm1
pmaddwd 40(%esi),%mm5
movq 16(%edx),%mm2
movq 80(%edx),%mm6
pmaddwd 16(%esi),%mm2
pmaddwd 48(%esi),%mm6
movq 24(%edx),%mm3
movq 88(%edx),%mm7
pmaddwd 24(%esi),%mm3
pmaddwd 56(%esi),%mm7
paddd %mm1,%mm0
paddd %mm5,%mm4
paddd %mm2,%mm0
paddd %mm6,%mm4
paddd %mm3,%mm0
paddd %mm7,%mm4
movq %mm0,%mm1
movq %mm4,%mm5
psrlq $32,%mm1
psrlq $32,%mm5
paddd %mm1,%mm0
paddd %mm5,%mm4
psrad $13,%mm0
psrad $13,%mm4
packssdw %mm0,%mm0
packssdw %mm4,%mm4
movq (%edi), %mm1
punpckldq %mm4, %mm0
pand one_null, %mm1
pand null_one, %mm0
por %mm0, %mm1
movq %mm1,(%edi)
leal 64(%esi),%esi
leal 128(%edx),%edx
leal 8(%edi),%edi
decl %ecx
jnz .L3
popl %ecx
andl $1, %ecx
jecxz .next_loop
movq (%edx),%mm0
pmaddwd (%esi),%mm0
movq 8(%edx),%mm1
pmaddwd 8(%esi),%mm1
movq 16(%edx),%mm2
pmaddwd 16(%esi),%mm2
movq 24(%edx),%mm3
pmaddwd 24(%esi),%mm3
paddd %mm1,%mm0
paddd %mm2,%mm0
paddd %mm3,%mm0
movq %mm0,%mm1
psrlq $32,%mm1
paddd %mm1,%mm0
psrad $13,%mm0
packssdw %mm0,%mm0
movd %mm0,%eax
movw %ax, (%edi)
leal 32(%esi),%esi
leal 64(%edx),%edx
leal 4(%edi),%edi
.next_loop:
subl $64,%esi
movl $7,%ecx
.align 16
.L4:
movq (%edx),%mm0
movq 64(%edx),%mm4
pmaddwd (%esi),%mm0
pmaddwd -32(%esi),%mm4
movq 8(%edx),%mm1
movq 72(%edx),%mm5
pmaddwd 8(%esi),%mm1
pmaddwd -24(%esi),%mm5
movq 16(%edx),%mm2
movq 80(%edx),%mm6
pmaddwd 16(%esi),%mm2
pmaddwd -16(%esi),%mm6
movq 24(%edx),%mm3
movq 88(%edx),%mm7
pmaddwd 24(%esi),%mm3
pmaddwd -8(%esi),%mm7
paddd %mm1,%mm0
paddd %mm5,%mm4
paddd %mm2,%mm0
paddd %mm6,%mm4
paddd %mm3,%mm0
paddd %mm7,%mm4
movq %mm0,%mm1
movq %mm4,%mm5
psrlq $32,%mm1
psrlq $32,%mm5
paddd %mm0,%mm1
paddd %mm4,%mm5
psrad $13,%mm1
psrad $13,%mm5
packssdw %mm1,%mm1
packssdw %mm5,%mm5
psubd %mm0,%mm0
psubd %mm4,%mm4
psubsw %mm1,%mm0
psubsw %mm5,%mm4
movq (%edi), %mm1
punpckldq %mm4, %mm0
pand one_null, %mm1
pand null_one, %mm0
por %mm0, %mm1
movq %mm1,(%edi)
subl $64,%esi
addl $128,%edx
leal 8(%edi),%edi
decl %ecx
jnz .L4
movq (%edx),%mm0
pmaddwd (%esi),%mm0
movq 8(%edx),%mm1
pmaddwd 8(%esi),%mm1
movq 16(%edx),%mm2
pmaddwd 16(%esi),%mm2
movq 24(%edx),%mm3
pmaddwd 24(%esi),%mm3
paddd %mm1,%mm0
paddd %mm2,%mm0
paddd %mm3,%mm0
movq %mm0,%mm1
psrlq $32,%mm1
paddd %mm0,%mm1
psrad $13,%mm1
packssdw %mm1,%mm1
psubd %mm0,%mm0
psubsw %mm1,%mm0
movd %mm0,%eax
movw %ax,(%edi)
emms
popl %ebx
popl %esi
popl %edi
popl %ebp
ret
|
