aboutsummaryrefslogtreecommitdiffstats
path: root/libswresample/x86/audio_convert.asm
blob: 5768405733648ed0e25a355335e7bdfaf2fdef35 (plain) (blame)
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
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
;******************************************************************************
;* 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

word_unpack_shuf : db  0, 1, 4, 5, 8, 9,12,13, 2, 3, 6, 7,10,11,14,15

SECTION .text


;to, from, a/u, log2_outsize, log_intsize, const
%macro PACK_2CH 5-7
cglobal pack_2ch_%2_to_%1_%3, 3, 4, 6, dst, src, len, src2
    mov src2q   , [srcq+gprsize]
    mov srcq    , [srcq]
    mov dstq    , [dstq]
%ifidn %3, a
    test dstq, mmsize-1
        jne pack_2ch_%2_to_%1_u_int %+ SUFFIX
    test srcq, mmsize-1
        jne pack_2ch_%2_to_%1_u_int %+ SUFFIX
    test src2q, mmsize-1
        jne pack_2ch_%2_to_%1_u_int %+ SUFFIX
%else
pack_2ch_%2_to_%1_u_int %+ SUFFIX
%endif
    lea     srcq , [srcq  + (1<<%5)*lenq]
    lea     src2q, [src2q + (1<<%5)*lenq]
    lea     dstq , [dstq  + (2<<%4)*lenq]
    neg     lenq
    %7
.next:
%if %4 >= %5
    mov%3     m0, [         srcq +(1<<%5)*lenq]
    mova      m1, m0
    mov%3     m2, [         src2q+(1<<%5)*lenq]
%if %5 == 1
    punpcklwd m0, m2
    punpckhwd m1, m2
%else
    punpckldq m0, m2
    punpckhdq m1, m2
%endif
    %6
%else
    mov%3     m0, [         srcq +(1<<%5)*lenq]
    mov%3     m1, [mmsize + srcq +(1<<%5)*lenq]
    mov%3     m2, [         src2q+(1<<%5)*lenq]
    mov%3     m3, [mmsize + src2q+(1<<%5)*lenq]
    %6
    mova      m2, m0
    punpcklwd m0, m1
    punpckhwd m2, m1
    SWAP 1,2
%endif
    mov%3 [           dstq+(2<<%4)*lenq], m0
    mov%3 [  mmsize + dstq+(2<<%4)*lenq], m1
%if %4 > %5
    mov%3 [2*mmsize + dstq+(2<<%4)*lenq], m2
    mov%3 [3*mmsize + dstq+(2<<%4)*lenq], m3
    add lenq, 4*mmsize/(2<<%4)
%else
    add lenq, 2*mmsize/(2<<%4)
%endif
        jl .next
    REP_RET
%endmacro

%macro UNPACK_2CH 5-7
cglobal unpack_2ch_%2_to_%1_%3, 3, 4, 7, dst, src, len, dst2
    mov dst2q   , [dstq+gprsize]
    mov srcq    , [srcq]
    mov dstq    , [dstq]
%ifidn %3, a
    test dstq, mmsize-1
        jne unpack_2ch_%2_to_%1_u_int %+ SUFFIX
    test srcq, mmsize-1
        jne unpack_2ch_%2_to_%1_u_int %+ SUFFIX
    test dst2q, mmsize-1
        jne unpack_2ch_%2_to_%1_u_int %+ SUFFIX
%else
unpack_2ch_%2_to_%1_u_int %+ SUFFIX
%endif
    lea     srcq , [srcq  + (2<<%5)*lenq]
    lea     dstq , [dstq  + (1<<%4)*lenq]
    lea     dst2q, [dst2q + (1<<%4)*lenq]
    neg     lenq
    %7
    mova      m6, [word_unpack_shuf]
.next:
    mov%3     m0, [           srcq +(2<<%5)*lenq]
    mov%3     m2, [  mmsize + srcq +(2<<%5)*lenq]
%if %5 == 1
%ifidn SUFFIX, _ssse3
    pshufb    m0, m6
    mova      m1, m0
    pshufb    m2, m6
    punpcklqdq m0,m2
    punpckhqdq m1,m2
%else
    mova      m1, m0
    punpcklwd m0,m2
    punpckhwd m1,m2

    mova      m2, m0
    punpcklwd m0,m1
    punpckhwd m2,m1

    mova      m1, m0
    punpcklwd m0,m2
    punpckhwd m1,m2
%endif
%else
    mova      m1, m0
    shufps    m0, m2, 10001000b
    shufps    m1, m2, 11011101b
%endif
%if %4 < %5
    mov%3     m2, [2*mmsize + srcq +(2<<%5)*lenq]
    mova      m3, m2
    mov%3     m4, [3*mmsize + srcq +(2<<%5)*lenq]
    shufps    m2, m4, 10001000b
    shufps    m3, m4, 11011101b
    SWAP 1,2
%endif
    %6
    mov%3 [           dstq+(1<<%4)*lenq], m0
%if %4 > %5
    mov%3 [          dst2q+(1<<%4)*lenq], m2
    mov%3 [ mmsize +  dstq+(1<<%4)*lenq], m1
    mov%3 [ mmsize + dst2q+(1<<%4)*lenq], m3
    add lenq, 2*mmsize/(1<<%4)
%else
    mov%3 [          dst2q+(1<<%4)*lenq], m1
    add lenq, mmsize/(1<<%4)
%endif
        jl .next
    REP_RET
%endmacro

%macro CONV 5-7
cglobal %2_to_%1_%3, 3, 3, 6, dst, src, len
    mov srcq    , [srcq]
    mov dstq    , [dstq]
%ifidn %3, a
    test dstq, mmsize-1
        jne %2_to_%1_u_int %+ SUFFIX
    test srcq, mmsize-1
        jne %2_to_%1_u_int %+ SUFFIX
%else
%2_to_%1_u_int %+ SUFFIX
%endif
    lea     srcq , [srcq  + (1<<%5)*lenq]
    lea     dstq , [dstq  + (1<<%4)*lenq]
    neg     lenq
    %7
.next:
    mov%3     m0, [           srcq +(1<<%5)*lenq]
    mov%3     m1, [  mmsize + srcq +(1<<%5)*lenq]
%if %4 < %5
    mov%3     m2, [2*mmsize + srcq +(1<<%5)*lenq]
    mov%3     m3, [3*mmsize + srcq +(1<<%5)*lenq]
%endif
    %6
    mov%3 [           dstq+(1<<%4)*lenq], m0
    mov%3 [  mmsize + dstq+(1<<%4)*lenq], m1
%if %4 > %5
    mov%3 [2*mmsize + dstq+(1<<%4)*lenq], m2
    mov%3 [3*mmsize + dstq+(1<<%4)*lenq], m3
    add lenq, 4*mmsize/(1<<%4)
%else
    add lenq, 2*mmsize/(1<<%4)
%endif
        jl .next
    REP_RET
%endmacro

%macro INT16_TO_INT32_N 0
    pxor      m2, m2
    pxor      m3, m3
    punpcklwd m2, m1
    punpckhwd m3, m1
    SWAP 4,0
    pxor      m0, m0
    pxor      m1, m1
    punpcklwd m0, m4
    punpckhwd m1, m4
%endmacro

%macro INT32_TO_INT16_N 0
    psrad     m0, 16
    psrad     m1, 16
    psrad     m2, 16
    psrad     m3, 16
    packssdw  m0, m1
    packssdw  m2, m3
    SWAP 1,2
%endmacro

%macro INT32_TO_FLOAT_INIT 0
    mova      m3, [flt2pm31]
%endmacro
%macro INT32_TO_FLOAT_N 0
    cvtdq2ps  m0, m0
    cvtdq2ps  m1, m1
    mulps m0, m0, m3
    mulps m1, m1, m3
%endmacro

%macro FLOAT_TO_INT32_INIT 0
    mova      m3, [flt2p31]
%endmacro
%macro FLOAT_TO_INT32_N 0
    mulps m0, m3
    mulps m1, m3
    cvtps2dq  m2, m0
    cvtps2dq  m4, m1
    cmpnltps m0, m3
    cmpnltps m1, m3
    paddd m0, m2
    paddd m1, m4
%endmacro

%macro INT16_TO_FLOAT_INIT 0
    mova      m5, [flt2pm31]
%endmacro
%macro INT16_TO_FLOAT_N 0
    INT16_TO_INT32_N
    cvtdq2ps  m0, m0
    cvtdq2ps  m1, m1
    cvtdq2ps  m2, m2
    cvtdq2ps  m3, m3
    mulps m0, m0, m5
    mulps m1, m1, m5
    mulps m2, m2, m5
    mulps m3, m3, m5
%endmacro

%macro FLOAT_TO_INT16_INIT 0
    mova      m5, [flt2p15]
%endmacro
%macro FLOAT_TO_INT16_N 0
    mulps m0, m5
    mulps m1, m5
    mulps m2, m5
    mulps m3, m5
    cvtps2dq  m0, m0
    cvtps2dq  m1, m1
    packssdw  m0, m1
    cvtps2dq  m1, m2
    cvtps2dq  m3, m3
    packssdw  m1, m3
%endmacro

INIT_MMX mmx
CONV int32, int16, u, 2, 1, INT16_TO_INT32_N
CONV int32, int16, a, 2, 1, INT16_TO_INT32_N
CONV int16, int32, u, 1, 2, INT32_TO_INT16_N
CONV int16, int32, a, 1, 2, INT32_TO_INT16_N

INIT_XMM sse
CONV int32, int16, u, 2, 1, INT16_TO_INT32_N
CONV int32, int16, a, 2, 1, INT16_TO_INT32_N
CONV int16, int32, u, 1, 2, INT32_TO_INT16_N
CONV int16, int32, a, 1, 2, INT32_TO_INT16_N

PACK_2CH int16, int16, u, 1, 1
PACK_2CH int16, int16, a, 1, 1
PACK_2CH int32, int32, u, 2, 2
PACK_2CH int32, int32, a, 2, 2
PACK_2CH int32, int16, u, 2, 1, INT16_TO_INT32_N
PACK_2CH int32, int16, a, 2, 1, INT16_TO_INT32_N
PACK_2CH int16, int32, u, 1, 2, INT32_TO_INT16_N
PACK_2CH int16, int32, a, 1, 2, INT32_TO_INT16_N

UNPACK_2CH int16, int16, u, 1, 1
UNPACK_2CH int16, int16, a, 1, 1
UNPACK_2CH int32, int32, u, 2, 2
UNPACK_2CH int32, int32, a, 2, 2
UNPACK_2CH int32, int16, u, 2, 1, INT16_TO_INT32_N
UNPACK_2CH int32, int16, a, 2, 1, INT16_TO_INT32_N
UNPACK_2CH int16, int32, u, 1, 2, INT32_TO_INT16_N
UNPACK_2CH int16, int32, a, 1, 2, INT32_TO_INT16_N

INIT_XMM sse2
CONV float, int32, u, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT
CONV float, int32, a, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT
CONV int32, float, u, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT
CONV int32, float, a, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT
CONV float, int16, u, 2, 1, INT16_TO_FLOAT_N, INT16_TO_FLOAT_INIT
CONV float, int16, a, 2, 1, INT16_TO_FLOAT_N, INT16_TO_FLOAT_INIT
CONV int16, float, u, 1, 2, FLOAT_TO_INT16_N, FLOAT_TO_INT16_INIT
CONV int16, float, a, 1, 2, FLOAT_TO_INT16_N, FLOAT_TO_INT16_INIT

PACK_2CH float, int32, u, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT
PACK_2CH float, int32, a, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT
PACK_2CH int32, float, u, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT
PACK_2CH int32, float, a, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT
PACK_2CH float, int16, u, 2, 1, INT16_TO_FLOAT_N, INT16_TO_FLOAT_INIT
PACK_2CH float, int16, a, 2, 1, INT16_TO_FLOAT_N, INT16_TO_FLOAT_INIT
PACK_2CH int16, float, u, 1, 2, FLOAT_TO_INT16_N, FLOAT_TO_INT16_INIT
PACK_2CH int16, float, a, 1, 2, FLOAT_TO_INT16_N, FLOAT_TO_INT16_INIT

UNPACK_2CH float, int32, u, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT
UNPACK_2CH float, int32, a, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT
UNPACK_2CH int32, float, u, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT
UNPACK_2CH int32, float, a, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT
UNPACK_2CH float, int16, u, 2, 1, INT16_TO_FLOAT_N, INT16_TO_FLOAT_INIT
UNPACK_2CH float, int16, a, 2, 1, INT16_TO_FLOAT_N, INT16_TO_FLOAT_INIT
UNPACK_2CH int16, float, u, 1, 2, FLOAT_TO_INT16_N, FLOAT_TO_INT16_INIT
UNPACK_2CH int16, float, a, 1, 2, FLOAT_TO_INT16_N, FLOAT_TO_INT16_INIT


INIT_XMM ssse3
UNPACK_2CH int16, int16, u, 1, 1
UNPACK_2CH int16, int16, a, 1, 1
UNPACK_2CH int32, int16, u, 2, 1, INT16_TO_INT32_N
UNPACK_2CH int32, int16, a, 2, 1, INT16_TO_INT32_N
UNPACK_2CH float, int16, u, 2, 1, INT16_TO_FLOAT_N, INT16_TO_FLOAT_INIT
UNPACK_2CH float, int16, a, 2, 1, INT16_TO_FLOAT_N, INT16_TO_FLOAT_INIT

%if HAVE_AVX
INIT_YMM avx
CONV float, int32, u, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT
CONV float, int32, a, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT
%endif