audio_convert.asm 20.3 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
;******************************************************************************
;* 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/x86util.asm"

23
SECTION_RODATA 32
24
flt2pm31: times 8 dd 4.6566129e-10
25
flt2p31 : times 8 dd 2147483648.0
26
flt2p15 : times 8 dd 32768.0
27

28 29
word_unpack_shuf : db  0, 1, 4, 5, 8, 9,12,13, 2, 3, 6, 7,10,11,14,15

30 31
SECTION .text

32

33 34
;to, from, a/u, log2_outsize, log_intsize, const
%macro PACK_2CH 5-7
35
cglobal pack_2ch_%2_to_%1_%3, 3, 4, 6, dst, src, len, src2
36 37 38 39 40
    mov src2q   , [srcq+gprsize]
    mov srcq    , [srcq]
    mov dstq    , [dstq]
%ifidn %3, a
    test dstq, mmsize-1
41
        jne pack_2ch_%2_to_%1_u_int %+ SUFFIX
42
    test srcq, mmsize-1
43
        jne pack_2ch_%2_to_%1_u_int %+ SUFFIX
44
    test src2q, mmsize-1
45
        jne pack_2ch_%2_to_%1_u_int %+ SUFFIX
46
%else
47
pack_2ch_%2_to_%1_u_int %+ SUFFIX
48 49 50 51 52
%endif
    lea     srcq , [srcq  + (1<<%5)*lenq]
    lea     src2q, [src2q + (1<<%5)*lenq]
    lea     dstq , [dstq  + (2<<%4)*lenq]
    neg     lenq
53
    %7 m0,m1,m2,m3,m4,m5
54
.next:
55
%if %4 >= %5
56 57 58 59 60 61 62 63 64 65
    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
66
    %6 m0,m1,m2,m3,m4,m5
67 68 69 70 71
%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]
72
    %6 m0,m1,m2,m3,m4,m5
73 74 75 76 77
    mova      m2, m0
    punpcklwd m0, m1
    punpckhwd m2, m1
    SWAP 1,2
%endif
78 79 80 81 82 83 84 85 86 87 88 89 90
    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

91
%macro UNPACK_2CH 5-7
92
cglobal unpack_2ch_%2_to_%1_%3, 3, 4, 7, dst, src, len, dst2
93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109
    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
110
    %7 m0,m1,m2,m3,m4,m5
111
    mova      m6, [word_unpack_shuf]
112 113 114 115
.next:
    mov%3     m0, [           srcq +(2<<%5)*lenq]
    mov%3     m2, [  mmsize + srcq +(2<<%5)*lenq]
%if %5 == 1
116 117 118 119 120 121 122 123
%ifidn SUFFIX, _ssse3
    pshufb    m0, m6
    mova      m1, m0
    pshufb    m2, m6
    punpcklqdq m0,m2
    punpckhqdq m1,m2
%else
    mova      m1, m0
124 125 126 127 128 129 130 131 132 133
    punpcklwd m0,m2
    punpckhwd m1,m2

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

    mova      m1, m0
    punpcklwd m0,m2
    punpckhwd m1,m2
134
%endif
135
%else
136
    mova      m1, m0
137 138 139 140 141 142 143 144 145 146 147
    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
148
    %6 m0,m1,m2,m3,m4,m5
149 150 151 152 153 154 155 156 157 158 159 160 161 162
    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

163 164 165 166 167 168 169 170 171 172 173 174 175 176 177
%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
178
    %7 m0,m1,m2,m3,m4,m5
179 180 181 182 183 184 185
.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
186
    %6 m0,m1,m2,m3,m4,m5
187 188 189 190 191 192 193 194 195 196
    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
197 198 199 200
%if mmsize == 8
    emms
    RET
%else
201
    REP_RET
202
%endif
203 204
%endmacro

205
%macro PACK_6CH 5-7
206 207 208 209 210 211 212 213 214 215 216 217 218
cglobal pack_6ch_%2_to_%1_%3, 2,8,7, dst, src, src1, src2, src3, src4, src5, len
%if ARCH_X86_64
    mov     lend, r2d
%else
    %define lend dword r2m
%endif
    mov    src1q, [srcq+1*gprsize]
    mov    src2q, [srcq+2*gprsize]
    mov    src3q, [srcq+3*gprsize]
    mov    src4q, [srcq+4*gprsize]
    mov    src5q, [srcq+5*gprsize]
    mov     srcq, [srcq]
    mov     dstq, [dstq]
219 220 221 222 223
%ifidn %3, a
    test dstq, mmsize-1
        jne pack_6ch_%2_to_%1_u_int %+ SUFFIX
    test srcq, mmsize-1
        jne pack_6ch_%2_to_%1_u_int %+ SUFFIX
224 225
    test src1q, mmsize-1
        jne pack_6ch_%2_to_%1_u_int %+ SUFFIX
226 227 228 229 230 231 232 233 234 235 236
    test src2q, mmsize-1
        jne pack_6ch_%2_to_%1_u_int %+ SUFFIX
    test src3q, mmsize-1
        jne pack_6ch_%2_to_%1_u_int %+ SUFFIX
    test src4q, mmsize-1
        jne pack_6ch_%2_to_%1_u_int %+ SUFFIX
    test src5q, mmsize-1
        jne pack_6ch_%2_to_%1_u_int %+ SUFFIX
%else
pack_6ch_%2_to_%1_u_int %+ SUFFIX
%endif
237 238 239 240 241
    sub    src1q, srcq
    sub    src2q, srcq
    sub    src3q, srcq
    sub    src4q, srcq
    sub    src5q, srcq
242
    %7 x,x,x,x,m7,x
243
.loop:
244 245 246 247 248 249
    mov%3     m0, [srcq      ]
    mov%3     m1, [srcq+src1q]
    mov%3     m2, [srcq+src2q]
    mov%3     m3, [srcq+src3q]
    mov%3     m4, [srcq+src4q]
    mov%3     m5, [srcq+src5q]
250
%if cpuflag(sse)
251 252 253 254
    SBUTTERFLYPS 0, 1, 6
    SBUTTERFLYPS 2, 3, 6
    SBUTTERFLYPS 4, 5, 6

255
%if cpuflag(avx)
256
    blendps   m6, m4, m0, 1100b
257 258 259 260 261
%else
    movaps    m6, m4
    shufps    m4, m0, q3210
    SWAP 4,6
%endif
262 263
    movlhps   m0, m2
    movhlps   m4, m2
264
%if cpuflag(avx)
265
    blendps   m2, m5, m1, 1100b
266 267 268 269 270
%else
    movaps    m2, m5
    shufps    m5, m1, q3210
    SWAP 2,5
%endif
271 272 273
    movlhps   m1, m3
    movhlps   m5, m3

274 275 276 277
    %6 m0,m6,x,x,m7,m3
    %6 m4,m1,x,x,m7,m3
    %6 m2,m5,x,x,m7,m3

278 279 280 281 282 283
    mov %+ %3 %+ ps [dstq   ], m0
    mov %+ %3 %+ ps [dstq+16], m6
    mov %+ %3 %+ ps [dstq+32], m4
    mov %+ %3 %+ ps [dstq+48], m1
    mov %+ %3 %+ ps [dstq+64], m2
    mov %+ %3 %+ ps [dstq+80], m5
284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307
%else ; mmx
    SBUTTERFLY dq, 0, 1, 6
    SBUTTERFLY dq, 2, 3, 6
    SBUTTERFLY dq, 4, 5, 6

    movq   [dstq   ], m0
    movq   [dstq+ 8], m2
    movq   [dstq+16], m4
    movq   [dstq+24], m1
    movq   [dstq+32], m3
    movq   [dstq+40], m5
%endif
    add      srcq, mmsize
    add      dstq, mmsize*6
    sub      lend, mmsize/4
    jg .loop
%if mmsize == 8
    emms
    RET
%else
    REP_RET
%endif
%endmacro

308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380
%macro UNPACK_6CH 5-7
cglobal unpack_6ch_%2_to_%1_%3, 2, 8, 8, dst, src, dst1, dst2, dst3, dst4, dst5, len
%if ARCH_X86_64
    mov     lend, r2d
%else
    %define lend dword r2m
%endif
    mov    dst1q, [dstq+1*gprsize]
    mov    dst2q, [dstq+2*gprsize]
    mov    dst3q, [dstq+3*gprsize]
    mov    dst4q, [dstq+4*gprsize]
    mov    dst5q, [dstq+5*gprsize]
    mov     dstq, [dstq]
    mov     srcq, [srcq]
%ifidn %3, a
    test dstq, mmsize-1
        jne unpack_6ch_%2_to_%1_u_int %+ SUFFIX
    test srcq, mmsize-1
        jne unpack_6ch_%2_to_%1_u_int %+ SUFFIX
    test dst1q, mmsize-1
        jne unpack_6ch_%2_to_%1_u_int %+ SUFFIX
    test dst2q, mmsize-1
        jne unpack_6ch_%2_to_%1_u_int %+ SUFFIX
    test dst3q, mmsize-1
        jne unpack_6ch_%2_to_%1_u_int %+ SUFFIX
    test dst4q, mmsize-1
        jne unpack_6ch_%2_to_%1_u_int %+ SUFFIX
    test dst5q, mmsize-1
        jne unpack_6ch_%2_to_%1_u_int %+ SUFFIX
%else
unpack_6ch_%2_to_%1_u_int %+ SUFFIX
%endif
    sub    dst1q, dstq
    sub    dst2q, dstq
    sub    dst3q, dstq
    sub    dst4q, dstq
    sub    dst5q, dstq
    %7 x,x,x,x,m7,x
.loop:
    mov%3     m0, [srcq   ]
    mov%3     m1, [srcq+16]
    mov%3     m2, [srcq+32]
    mov%3     m3, [srcq+48]
    mov%3     m4, [srcq+64]
    mov%3     m5, [srcq+80]

    SBUTTERFLYPS 0, 3, 6
    SBUTTERFLYPS 1, 4, 6
    SBUTTERFLYPS 2, 5, 6
    SBUTTERFLYPS 0, 4, 6
    SBUTTERFLYPS 3, 2, 6
    SBUTTERFLYPS 1, 5, 6
    SWAP 1, 4
    SWAP 2, 3

    %6 m0,m1,x,x,m7,m6
    %6 m2,m3,x,x,m7,m6
    %6 m4,m5,x,x,m7,m6

    mov %+ %3 %+ ps [dstq      ], m0
    mov %+ %3 %+ ps [dstq+dst1q], m1
    mov %+ %3 %+ ps [dstq+dst2q], m2
    mov %+ %3 %+ ps [dstq+dst3q], m3
    mov %+ %3 %+ ps [dstq+dst4q], m4
    mov %+ %3 %+ ps [dstq+dst5q], m5

    add      srcq, mmsize*6
    add      dstq, mmsize
    sub      lend, mmsize/4
    jg .loop
    REP_RET
%endmacro

381 382
%define PACK_8CH_GPRS (10 * ARCH_X86_64) + ((6 + HAVE_ALIGNED_STACK) * ARCH_X86_32)

383
%macro PACK_8CH 5-7
384
cglobal pack_8ch_%2_to_%1_%3, 2,PACK_8CH_GPRS,10, ARCH_X86_32*48, dst, src, len, src1, src2, src3, src4, src5, src6, src7
385 386 387 388 389
    mov     dstq, [dstq]
%if ARCH_X86_32
    DEFINE_ARGS dst, src, src2, src3, src4, src5, src6
    %define lend dword r2m
    %define src1q r0q
390 391 392 393 394 395
    %define src1m dword [rsp+32]
%if HAVE_ALIGNED_STACK == 0
    DEFINE_ARGS dst, src, src2, src3, src5, src6
    %define src4q r0q
    %define src4m dword [rsp+36]
%endif
396
    %define src7q r0q
397
    %define src7m dword [rsp+40]
398 399 400 401 402
    mov     dstm, dstq
%endif
    mov    src7q, [srcq+7*gprsize]
    mov    src6q, [srcq+6*gprsize]
%if ARCH_X86_32
403
    mov    src7m, src7q
404
%endif
405 406
    mov    src5q, [srcq+5*gprsize]
    mov    src4q, [srcq+4*gprsize]
407
    mov    src3q, [srcq+3*gprsize]
408 409 410
%if ARCH_X86_32 && HAVE_ALIGNED_STACK == 0
    mov    src4m, src4q
%endif
411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428
    mov    src2q, [srcq+2*gprsize]
    mov    src1q, [srcq+1*gprsize]
    mov     srcq, [srcq]
%ifidn %3, a
%if ARCH_X86_32
    test dstmp, mmsize-1
%else
    test dstq, mmsize-1
%endif
        jne pack_8ch_%2_to_%1_u_int %+ SUFFIX
    test srcq, mmsize-1
        jne pack_8ch_%2_to_%1_u_int %+ SUFFIX
    test src1q, mmsize-1
        jne pack_8ch_%2_to_%1_u_int %+ SUFFIX
    test src2q, mmsize-1
        jne pack_8ch_%2_to_%1_u_int %+ SUFFIX
    test src3q, mmsize-1
        jne pack_8ch_%2_to_%1_u_int %+ SUFFIX
429 430 431
%if ARCH_X86_32 && HAVE_ALIGNED_STACK == 0
    test src4m, mmsize-1
%else
432
    test src4q, mmsize-1
433
%endif
434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450
        jne pack_8ch_%2_to_%1_u_int %+ SUFFIX
    test src5q, mmsize-1
        jne pack_8ch_%2_to_%1_u_int %+ SUFFIX
    test src6q, mmsize-1
        jne pack_8ch_%2_to_%1_u_int %+ SUFFIX
%if ARCH_X86_32
    test src7m, mmsize-1
%else
    test src7q, mmsize-1
%endif
        jne pack_8ch_%2_to_%1_u_int %+ SUFFIX
%else
pack_8ch_%2_to_%1_u_int %+ SUFFIX
%endif
    sub    src1q, srcq
    sub    src2q, srcq
    sub    src3q, srcq
451
%if ARCH_X86_64 || HAVE_ALIGNED_STACK
452
    sub    src4q, srcq
453 454 455
%else
    sub    src4m, srcq
%endif
456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476
    sub    src5q, srcq
    sub    src6q, srcq
%if ARCH_X86_64
    sub    src7q, srcq
%else
    mov src1m, src1q
    sub src7m, srcq
%endif

%if ARCH_X86_64
    %7 x,x,x,x,m9,x
%elifidn %1, int32
    %define m9 [flt2p31]
%else
    %define m9 [flt2pm31]
%endif

.loop:
    mov%3     m0, [srcq      ]
    mov%3     m1, [srcq+src1q]
    mov%3     m2, [srcq+src2q]
477 478 479
%if ARCH_X86_32 && HAVE_ALIGNED_STACK == 0
    mov    src4q, src4m
%endif
480
    mov%3     m3, [srcq+src3q]
481 482
    mov%3     m4, [srcq+src4q]
    mov%3     m5, [srcq+src5q]
483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530
%if ARCH_X86_32
    mov    src7q, src7m
%endif
    mov%3     m6, [srcq+src6q]
    mov%3     m7, [srcq+src7q]

%if ARCH_X86_64
    TRANSPOSE8x4D 0, 1, 2, 3, 4, 5, 6, 7, 8

    %6 m0,m1,x,x,m9,m8
    %6 m2,m3,x,x,m9,m8
    %6 m4,m5,x,x,m9,m8
    %6 m6,m7,x,x,m9,m8

    mov%3 [dstq], m0
%else
    mov     dstq, dstm

    TRANSPOSE8x4D 0, 1, 2, 3, 4, 5, 6, 7, [rsp], [rsp+16], 1

    %6 m0,m1,x,x,m9,m2
    mova     m2, [rsp]
    mov%3   [dstq], m0
    %6 m2,m3,x,x,m9,m0
    %6 m4,m5,x,x,m9,m0
    %6 m6,m7,x,x,m9,m0

%endif

    mov%3 [dstq+16],  m1
    mov%3 [dstq+32],  m2
    mov%3 [dstq+48],  m3
    mov%3 [dstq+64],  m4
    mov%3 [dstq+80],  m5
    mov%3 [dstq+96],  m6
    mov%3 [dstq+112], m7

    add      srcq, mmsize
    add      dstq, mmsize*8
%if ARCH_X86_32
    mov      dstm, dstq
    mov      src1q, src1m
%endif
    sub      lend, mmsize/4
    jg .loop
    REP_RET
%endmacro

531
%macro INT16_TO_INT32_N 6
532 533 534 535 536 537 538 539 540 541 542
    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

543
%macro INT32_TO_INT16_N 6
544 545 546 547 548 549 550 551 552
    psrad     m0, 16
    psrad     m1, 16
    psrad     m2, 16
    psrad     m3, 16
    packssdw  m0, m1
    packssdw  m2, m3
    SWAP 1,2
%endmacro

553 554
%macro INT32_TO_FLOAT_INIT 6
    mova      %5, [flt2pm31]
555
%endmacro
556 557 558 559 560
%macro INT32_TO_FLOAT_N 6
    cvtdq2ps  %1, %1
    cvtdq2ps  %2, %2
    mulps %1, %1, %5
    mulps %2, %2, %5
561 562
%endmacro

563 564
%macro FLOAT_TO_INT32_INIT 6
    mova      %5, [flt2p31]
565
%endmacro
566 567 568
%macro FLOAT_TO_INT32_N 6
    mulps %1, %5
    mulps %2, %5
569
    cvtps2dq  %6, %1
570
    cmpps %1, %1, %5, 5
571 572
    paddd %1, %6
    cvtps2dq  %6, %2
573
    cmpps %2, %2, %5, 5
574
    paddd %2, %6
575 576
%endmacro

577
%macro INT16_TO_FLOAT_INIT 6
578 579
    mova      m5, [flt2pm31]
%endmacro
580 581
%macro INT16_TO_FLOAT_N 6
    INT16_TO_INT32_N %1,%2,%3,%4,%5,%6
582 583 584 585 586 587 588 589 590 591
    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

592
%macro FLOAT_TO_INT16_INIT 6
593 594
    mova      m5, [flt2p15]
%endmacro
595
%macro FLOAT_TO_INT16_N 6
596 597 598 599 600 601 602 603 604 605 606
    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
607

608
%macro NOP_N 0-6
609 610
%endmacro

611
INIT_MMX mmx
612 613 614 615
CONV int32, int16, u, 2, 1, INT16_TO_INT32_N, NOP_N
CONV int32, int16, a, 2, 1, INT16_TO_INT32_N, NOP_N
CONV int16, int32, u, 1, 2, INT32_TO_INT16_N, NOP_N
CONV int16, int32, a, 1, 2, INT32_TO_INT16_N, NOP_N
616

617 618
PACK_6CH float, float, u, 2, 2, NOP_N, NOP_N
PACK_6CH float, float, a, 2, 2, NOP_N, NOP_N
619

620 621 622 623
INIT_XMM sse
PACK_6CH float, float, u, 2, 2, NOP_N, NOP_N
PACK_6CH float, float, a, 2, 2, NOP_N, NOP_N

624 625 626
UNPACK_6CH float, float, u, 2, 2, NOP_N, NOP_N
UNPACK_6CH float, float, a, 2, 2, NOP_N, NOP_N

627
INIT_XMM sse2
628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649
CONV int32, int16, u, 2, 1, INT16_TO_INT32_N, NOP_N
CONV int32, int16, a, 2, 1, INT16_TO_INT32_N, NOP_N
CONV int16, int32, u, 1, 2, INT32_TO_INT16_N, NOP_N
CONV int16, int32, a, 1, 2, INT32_TO_INT16_N, NOP_N

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

UNPACK_2CH int16, int16, u, 1, 1, NOP_N, NOP_N
UNPACK_2CH int16, int16, a, 1, 1, NOP_N, NOP_N
UNPACK_2CH int32, int32, u, 2, 2, NOP_N, NOP_N
UNPACK_2CH int32, int32, a, 2, 2, NOP_N, NOP_N
UNPACK_2CH int32, int16, u, 2, 1, INT16_TO_INT32_N, NOP_N
UNPACK_2CH int32, int16, a, 2, 1, INT16_TO_INT32_N, NOP_N
UNPACK_2CH int16, int32, u, 1, 2, INT32_TO_INT16_N, NOP_N
UNPACK_2CH int16, int32, a, 1, 2, INT32_TO_INT16_N, NOP_N
650

651 652 653 654 655 656 657 658
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
659

660 661 662 663 664 665 666 667 668
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

669 670 671 672 673 674 675 676
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
677

678 679 680 681
PACK_6CH float, int32, u, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT
PACK_6CH float, int32, a, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT
PACK_6CH int32, float, u, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT
PACK_6CH int32, float, a, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT
682

683 684 685 686 687
UNPACK_6CH float, int32, u, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT
UNPACK_6CH float, int32, a, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT
UNPACK_6CH int32, float, u, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT
UNPACK_6CH int32, float, a, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT

688 689 690 691 692 693 694 695
PACK_8CH float, float, u, 2, 2, NOP_N, NOP_N
PACK_8CH float, float, a, 2, 2, NOP_N, NOP_N

PACK_8CH float, int32, u, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT
PACK_8CH float, int32, a, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT
PACK_8CH int32, float, u, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT
PACK_8CH int32, float, a, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT

696
INIT_XMM ssse3
697 698 699 700
UNPACK_2CH int16, int16, u, 1, 1, NOP_N, NOP_N
UNPACK_2CH int16, int16, a, 1, 1, NOP_N, NOP_N
UNPACK_2CH int32, int16, u, 2, 1, INT16_TO_INT32_N, NOP_N
UNPACK_2CH int32, int16, a, 2, 1, INT16_TO_INT32_N, NOP_N
701 702 703
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

704
%if HAVE_AVX_EXTERNAL
705
INIT_XMM avx
706 707 708
PACK_6CH float, float, u, 2, 2, NOP_N, NOP_N
PACK_6CH float, float, a, 2, 2, NOP_N, NOP_N

709 710 711
UNPACK_6CH float, float, u, 2, 2, NOP_N, NOP_N
UNPACK_6CH float, float, a, 2, 2, NOP_N, NOP_N

712 713
PACK_6CH float, int32, u, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT
PACK_6CH float, int32, a, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT
714 715
PACK_6CH int32, float, u, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT
PACK_6CH int32, float, a, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT
716

717 718 719 720 721
UNPACK_6CH float, int32, u, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT
UNPACK_6CH float, int32, a, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT
UNPACK_6CH int32, float, u, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT
UNPACK_6CH int32, float, a, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT

722 723 724 725 726 727 728 729
PACK_8CH float, float, u, 2, 2, NOP_N, NOP_N
PACK_8CH float, float, a, 2, 2, NOP_N, NOP_N

PACK_8CH float, int32, u, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT
PACK_8CH float, int32, a, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT
PACK_8CH int32, float, u, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT
PACK_8CH int32, float, a, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT

730
INIT_YMM avx
731 732
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
733
%endif
734 735 736 737 738 739

%if HAVE_AVX2_EXTERNAL
INIT_YMM avx2
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
%endif