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
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
|
;******************************************************************************
;* Copyright (c) 2012 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/x86inc.asm"
%include "libavutil/x86/x86util.asm"
SECTION_RODATA
align 32
flt2pm31: times 8 dd 4.6566129e-10
flt2p31 : times 8 dd 2147483648.0
flt2p15 : times 8 dd 32768.0
SECTION .text
%macro INT16_TO_INT32 1
cglobal int16_to_int32_%1, 3, 3, 3, dst, src, len
mov srcq, [srcq]
mov dstq, [dstq]
%ifidn %1, a
test dstq, mmsize-1
jne int16_to_int32_u_int %+ SUFFIX
test srcq, mmsize-1
jne int16_to_int32_u_int %+ SUFFIX
%else
int16_to_int32_u_int %+ SUFFIX
%endif
add dstq, lenq
shr lenq, 1
add srcq, lenq
neg lenq
.next
mov%1 m2, [srcq+lenq]
pxor m0, m0
pxor m1, m1
punpcklwd m0, m2
punpckhwd m1, m2
mov%1 [ dstq+2*lenq], m0
mov%1 [mmsize + dstq+2*lenq], m1
add lenq, mmsize
jl .next
%if mmsize == 8
emms
%endif
REP_RET
%endmacro
%macro INT32_TO_FLOAT 1
cglobal int32_to_float_%1, 3, 3, 3, dst, src, len
mov srcq, [srcq]
mov dstq, [dstq]
%ifidn %1, a
test dstq, mmsize-1
jne int32_to_float_u_int %+ SUFFIX
test srcq, mmsize-1
jne int32_to_float_u_int %+ SUFFIX
%else
int32_to_float_u_int %+ SUFFIX
%endif
add srcq, lenq
add dstq, lenq
neg lenq
mova m2, [flt2pm31]
.next:
%ifidn %1, a
cvtdq2ps m0, [ srcq+lenq]
cvtdq2ps m1, [mmsize + srcq+lenq]
%else
movu m0, [ srcq+lenq]
movu m1, [mmsize + srcq+lenq]
cvtdq2ps m0, m0
cvtdq2ps m1, m1
%endif
mulps m0, m0, m2
mulps m1, m1, m2
mov%1 [ dstq+lenq], m0
mov%1 [mmsize + dstq+lenq], m1
add lenq, 2*mmsize
jl .next
REP_RET
%endmacro
%macro INT16_TO_FLOAT 1
cglobal int16_to_float_%1, 3, 3, 4, dst, src, len
mov srcq, [srcq]
mov dstq, [dstq]
%ifidn %1, a
test dstq, mmsize-1
jne int16_to_float_u_int %+ SUFFIX
test srcq, mmsize-1
jne int16_to_float_u_int %+ SUFFIX
%else
int16_to_float_u_int %+ SUFFIX
%endif
add dstq, lenq
shr lenq, 1
add srcq, lenq
neg lenq
mova m3, [flt2pm31]
.next:
mov%1 m2, [srcq+lenq]
pxor m0, m0
pxor m1, m1
punpcklwd m0, m2
punpckhwd m1, m2
cvtdq2ps m0, m0
cvtdq2ps m1, m1
mulps m0, m3
mulps m1, m3
mov%1 [ dstq+2*lenq], m0
mov%1 [mmsize + dstq+2*lenq], m1
add lenq, mmsize
jl .next
REP_RET
%endmacro
%macro FLOAT_TO_INT32 1
cglobal float_to_int32_%1, 3, 3, 5, dst, src, len
mov srcq, [srcq]
mov dstq, [dstq]
%ifidn %1, a
test dstq, mmsize-1
jne float_to_int32_u_int %+ SUFFIX
test srcq, mmsize-1
jne float_to_int32_u_int %+ SUFFIX
%else
float_to_int32_u_int %+ SUFFIX
%endif
add srcq, lenq
add dstq, lenq
neg lenq
mova m2, [flt2p31]
.next:
mov%1 m0, [ srcq+lenq]
mov%1 m1, [mmsize + srcq+lenq]
mulps m0, m2
mulps m1, m2
cvtps2dq m3, m0
cvtps2dq m4, m1
cmpnltps m0, m2
cmpnltps m1, m2
paddd m0, m3
paddd m1, m4
mov%1 [ dstq+lenq], m0
mov%1 [mmsize + dstq+lenq], m1
add lenq, 2*mmsize
jl .next
REP_RET
%endmacro
%macro FLOAT_TO_INT16 1
cglobal float_to_int16_%1, 3, 3, 3, dst, src, len
mov srcq, [srcq]
mov dstq, [dstq]
%ifidn %1, a
test dstq, mmsize-1
jne float_to_int16_u_int %+ SUFFIX
test srcq, mmsize-1
jne float_to_int16_u_int %+ SUFFIX
%else
float_to_int16_u_int %+ SUFFIX
%endif
lea srcq, [srcq + 2*lenq]
add dstq, lenq
neg lenq
mova m2, [flt2p15]
.next:
mov%1 m0, [ srcq+2*lenq]
mov%1 m1, [mmsize + srcq+2*lenq]
mulps m0, m2
mulps m1, m2
cvtps2dq m0, m0
cvtps2dq m1, m1
packssdw m0, m1
mov%1 [ dstq+lenq], m0
add lenq, mmsize
jl .next
REP_RET
%endmacro
%macro INT32_TO_INT16 1
cglobal int32_to_int16_%1, 3, 3, 2, dst, src, len
mov srcq, [srcq]
mov dstq, [dstq]
%ifidn %1, a
test dstq, mmsize-1
jne int32_to_int16_u_int %+ SUFFIX
test srcq, mmsize-1
jne int32_to_int16_u_int %+ SUFFIX
%else
int32_to_int16_u_int %+ SUFFIX
%endif
lea srcq, [srcq + 2*lenq]
add dstq, lenq
neg lenq
.next:
mov%1 m0, [ srcq+2*lenq]
mov%1 m1, [mmsize + srcq+2*lenq]
psrad m0, 16
psrad m1, 16
packssdw m0, m1
mov%1 [ dstq+lenq], m0
add lenq, mmsize
jl .next
REP_RET
%endmacro
INIT_MMX mmx
INT16_TO_INT32 u
INT16_TO_INT32 a
INT32_TO_INT16 u
INT32_TO_INT16 a
INIT_XMM sse
INT16_TO_INT32 u
INT16_TO_INT32 a
INT32_TO_INT16 u
INT32_TO_INT16 a
INIT_XMM sse2
INT32_TO_FLOAT u
INT32_TO_FLOAT a
INT16_TO_FLOAT u
INT16_TO_FLOAT a
FLOAT_TO_INT32 u
FLOAT_TO_INT32 a
FLOAT_TO_INT16 u
FLOAT_TO_INT16 a
%if HAVE_AVX
INIT_YMM avx
INT32_TO_FLOAT u
INT32_TO_FLOAT a
%endif
|
