aboutsummaryrefslogtreecommitdiffstats
path: root/libavcodec/x86/cfhddsp.asm
blob: 87c2df634ad5be11f466270f944bcc7dae994c3b (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
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
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
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
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
;******************************************************************************
;* x86-optimized functions for the CFHD decoder
;* Copyright (c) 2020 Paul B Mahol
;*
;* 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

factor_p1_n1: dw 1, -1, 1, -1, 1, -1, 1, -1,
factor_n1_p1: dw -1, 1, -1, 1, -1, 1, -1, 1,
factor_p11_n4: dw 11, -4, 11, -4, 11, -4, 11, -4,
factor_p5_p4: dw 5, 4, 5, 4, 5, 4, 5, 4,
pd_4: times 4 dd 4
pw_1: times 8 dw 1
pw_0: times 8 dw 0
pw_1023: times 8 dw 1023
pw_4095: times 8 dw 4095

SECTION .text

%macro CFHD_HORIZ_FILTER 1
%if %1 == 1023
cglobal cfhd_horiz_filter_clip10, 5, 6, 8 + 4 * ARCH_X86_64, output, low, high, width, x, temp
    shl        widthd, 1
%define ostrideq widthq
%define lwidthq  widthq
%define hwidthq  widthq
%elif %1 == 4095
cglobal cfhd_horiz_filter_clip12, 5, 6, 8 + 4 * ARCH_X86_64, output, low, high, width, x, temp
    shl        widthd, 1
%define ostrideq widthq
%define lwidthq  widthq
%define hwidthq  widthq
%else
%if ARCH_X86_64
cglobal cfhd_horiz_filter, 8, 11, 12, output, ostride, low, lwidth, high, hwidth, width, height, x, y, temp
    shl  ostrided, 1
    shl   lwidthd, 1
    shl   hwidthd, 1
    shl    widthd, 1

    mov        yd, heightd
    neg        yq
%else
cglobal cfhd_horiz_filter, 7, 7, 8, output, x, low, y, high, temp, width, height
    shl        xd, 1
    shl        yd, 1
    shl     tempd, 1
    shl    widthd, 1

    mov       xmp, xq
    mov       ymp, yq
    mov    tempmp, tempq

    mov        yd, r7m
    neg        yq

%define ostrideq xm
%define lwidthq  ym
%define hwidthq  tempm
%endif
%endif

%if ARCH_X86_64
    mova       m8, [factor_p1_n1]
    mova       m9, [factor_n1_p1]
    mova      m10, [pw_1]
    mova      m11, [pd_4]
%endif

%if %1 == 0
.looph:
%endif
    movsx          xq, word [lowq]
    imul           xq, 11

    movsx       tempq, word [lowq + 2]
    imul        tempq, -4
    add         tempq, xq

    movsx          xq, word [lowq + 4]
    add         tempq, xq
    add         tempq, 4
    sar         tempq, 3

    movsx          xq, word [highq]
    add         tempq, xq
    sar         tempq, 1

%if %1
    movd          xm0, tempd
    CLIPW          m0, [pw_0], [pw_%1]
    pextrw      tempd, xm0, 0
%endif
    mov  word [outputq], tempw

    movsx          xq, word [lowq]
    imul           xq, 5

    movsx       tempq, word [lowq + 2]
    imul        tempq, 4
    add         tempq, xq

    movsx          xq, word [lowq + 4]
    sub         tempq, xq
    add         tempq, 4
    sar         tempq, 3

    movsx          xq, word [highq]
    sub         tempq, xq
    sar         tempq, 1

%if %1
    movd          xm0, tempd
    CLIPW          m0, [pw_0], [pw_%1]
    pextrw      tempd, xm0, 0
%endif
    mov  word [outputq + 2], tempw

    mov            xq, 0

.loop:
    movu           m4, [lowq + xq]
    movu           m1, [lowq + xq + 4]

    mova           m5, m4
    punpcklwd      m4, m1
    punpckhwd      m5, m1

    mova           m6, m4
    mova           m7, m5

%if ARCH_X86_64
    pmaddwd        m4, m8
    pmaddwd        m5, m8
    pmaddwd        m6, m9
    pmaddwd        m7, m9

    paddd          m4, m11
    paddd          m5, m11
    paddd          m6, m11
    paddd          m7, m11
%else
    pmaddwd        m4, [factor_p1_n1]
    pmaddwd        m5, [factor_p1_n1]
    pmaddwd        m6, [factor_n1_p1]
    pmaddwd        m7, [factor_n1_p1]

    paddd          m4, [pd_4]
    paddd          m5, [pd_4]
    paddd          m6, [pd_4]
    paddd          m7, [pd_4]
%endif

    psrad          m4, 3
    psrad          m5, 3
    psrad          m6, 3
    psrad          m7, 3

    movu           m2, [lowq + xq + 2]
    movu           m3, [highq + xq + 2]

    mova           m0, m2
    punpcklwd      m2, m3
    punpckhwd      m0, m3

    mova           m1, m2
    mova           m3, m0

%if ARCH_X86_64
    pmaddwd        m2, m10
    pmaddwd        m0, m10
    pmaddwd        m1, m8
    pmaddwd        m3, m8
%else
    pmaddwd        m2, [pw_1]
    pmaddwd        m0, [pw_1]
    pmaddwd        m1, [factor_p1_n1]
    pmaddwd        m3, [factor_p1_n1]
%endif

    paddd          m2, m4
    paddd          m0, m5
    paddd          m1, m6
    paddd          m3, m7

    psrad          m2, 1
    psrad          m0, 1
    psrad          m1, 1
    psrad          m3, 1

    packssdw       m2, m0
    packssdw       m1, m3

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

%if %1
    CLIPW          m2, [pw_0], [pw_%1]
    CLIPW          m0, [pw_0], [pw_%1]
%endif

    movu  [outputq + xq * 2 + 4], m2
    movu  [outputq + xq * 2 + mmsize + 4], m0

    add            xq, mmsize
    cmp            xq, widthq
    jl .loop

    add          lowq, widthq
    add         highq, widthq
    add       outputq, widthq
    add       outputq, widthq

    movsx          xq, word [lowq - 2]
    imul           xq, 5

    movsx       tempq, word [lowq - 4]
    imul        tempq, 4
    add         tempq, xq

    movsx          xq, word [lowq - 6]
    sub         tempq, xq
    add         tempq, 4
    sar         tempq, 3

    movsx          xq, word [highq - 2]
    add         tempq, xq
    sar         tempq, 1

%if %1
    movd          xm0, tempd
    CLIPW          m0, [pw_0], [pw_%1]
    pextrw      tempd, xm0, 0
%endif
    mov  word [outputq - 4], tempw

    movsx          xq, word [lowq - 2]
    imul           xq, 11

    movsx       tempq, word [lowq - 4]
    imul        tempq, -4
    add         tempq, xq

    movsx          xq, word [lowq - 6]
    add         tempq, xq
    add         tempq, 4
    sar         tempq, 3

    movsx          xq, word [highq - 2]
    sub         tempq, xq
    sar         tempq, 1

%if %1
    movd          xm0, tempd
    CLIPW          m0, [pw_0], [pw_%1]
    pextrw      tempd, xm0, 0
%endif
    mov  word [outputq - 2], tempw

%if %1 == 0
    sub          lowq, widthq
    sub         highq, widthq
    sub       outputq, widthq
    sub       outputq, widthq

    add          lowq, lwidthq
    add         highq, hwidthq
    add       outputq, ostrideq
    add       outputq, ostrideq
    add            yq, 1
    jl .looph
%endif

    RET
%endmacro

INIT_XMM sse2
CFHD_HORIZ_FILTER 0

INIT_XMM sse2
CFHD_HORIZ_FILTER 1023

INIT_XMM sse2
CFHD_HORIZ_FILTER 4095

INIT_XMM sse2
%if ARCH_X86_64
cglobal cfhd_vert_filter, 8, 11, 14, output, ostride, low, lwidth, high, hwidth, width, height, x, y, pos
    shl        ostrided, 1
    shl         lwidthd, 1
    shl         hwidthd, 1
    shl          widthd, 1

    dec   heightd

    mova       m8, [factor_p1_n1]
    mova       m9, [factor_n1_p1]
    mova      m10, [pw_1]
    mova      m11, [pd_4]
    mova      m12, [factor_p11_n4]
    mova      m13, [factor_p5_p4]
%else
cglobal cfhd_vert_filter, 7, 7, 8, output, x, low, y, high, pos, width, height
    shl        xd, 1
    shl        yd, 1
    shl      posd, 1
    shl    widthd, 1

    mov       xmp, xq
    mov       ymp, yq
    mov     posmp, posq

    mov        xq, r7m
    dec        xq
    mov   widthmp, xq

%define ostrideq xm
%define lwidthq  ym
%define hwidthq  posm
%define heightq  widthm

%endif

    xor        xq, xq
.loopw:
    xor        yq, yq

    mov      posq, xq
    movu       m0, [lowq + posq]
    add      posq, lwidthq
    movu       m1, [lowq + posq]
    mova       m2, m0
    punpcklwd  m0, m1
    punpckhwd  m2, m1

%if ARCH_X86_64
    pmaddwd    m0, m12
    pmaddwd    m2, m12
%else
    pmaddwd    m0, [factor_p11_n4]
    pmaddwd    m2, [factor_p11_n4]
%endif

    pxor       m4, m4
    add      posq, lwidthq
    movu       m1, [lowq + posq]
    mova       m3, m4
    punpcklwd  m4, m1
    punpckhwd  m3, m1

    psrad      m4, 16
    psrad      m3, 16

    paddd      m0, m4
    paddd      m2, m3

    paddd      m0, [pd_4]
    paddd      m2, [pd_4]

    psrad      m0, 3
    psrad      m2, 3

    mov      posq, xq
    pxor       m4, m4
    movu       m1, [highq + posq]
    mova       m3, m4
    punpcklwd  m4, m1
    punpckhwd  m3, m1

    psrad      m4, 16
    psrad      m3, 16

    paddd      m0, m4
    paddd      m2, m3

    psrad      m0, 1
    psrad      m2, 1

    packssdw   m0, m2

    movu    [outputq + posq], m0

    movu       m0, [lowq + posq]
    add      posq, lwidthq
    movu       m1, [lowq + posq]
    mova       m2, m0
    punpcklwd  m0, m1
    punpckhwd  m2, m1

%if ARCH_X86_64
    pmaddwd    m0, m13
    pmaddwd    m2, m13
%else
    pmaddwd    m0, [factor_p5_p4]
    pmaddwd    m2, [factor_p5_p4]
%endif

    pxor       m4, m4
    add      posq, lwidthq
    movu       m1, [lowq + posq]
    mova       m3, m4
    punpcklwd  m4, m1
    punpckhwd  m3, m1

    psrad      m4, 16
    psrad      m3, 16

    psubd      m0, m4
    psubd      m2, m3

    paddd      m0, [pd_4]
    paddd      m2, [pd_4]

    psrad      m0, 3
    psrad      m2, 3

    mov      posq, xq
    pxor       m4, m4
    movu       m1, [highq + posq]
    mova       m3, m4
    punpcklwd  m4, m1
    punpckhwd  m3, m1

    psrad      m4, 16
    psrad      m3, 16

    psubd      m0, m4
    psubd      m2, m3

    psrad      m0, 1
    psrad      m2, 1

    packssdw   m0, m2

    add      posq, ostrideq
    movu    [outputq + posq], m0

    add        yq, 1
.looph:
    mov      posq, lwidthq
    imul     posq, yq
    sub      posq, lwidthq
    add      posq, xq

    movu       m4, [lowq + posq]

    add      posq, lwidthq
    add      posq, lwidthq
    movu       m1, [lowq + posq]

    mova       m5, m4
    punpcklwd  m4, m1
    punpckhwd  m5, m1

    mova       m6, m4
    mova       m7, m5

%if ARCH_X86_64
    pmaddwd    m4, m8
    pmaddwd    m5, m8
    pmaddwd    m6, m9
    pmaddwd    m7, m9

    paddd      m4, m11
    paddd      m5, m11
    paddd      m6, m11
    paddd      m7, m11
%else
    pmaddwd    m4, [factor_p1_n1]
    pmaddwd    m5, [factor_p1_n1]
    pmaddwd    m6, [factor_n1_p1]
    pmaddwd    m7, [factor_n1_p1]

    paddd      m4, [pd_4]
    paddd      m5, [pd_4]
    paddd      m6, [pd_4]
    paddd      m7, [pd_4]
%endif

    psrad      m4, 3
    psrad      m5, 3
    psrad      m6, 3
    psrad      m7, 3

    sub      posq, lwidthq
    movu       m0, [lowq + posq]

    mov      posq, hwidthq
    imul     posq, yq
    add      posq, xq
    movu       m1, [highq + posq]

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

    mova       m1, m0
    mova       m3, m2

%if ARCH_X86_64
    pmaddwd    m0, m10
    pmaddwd    m2, m10
    pmaddwd    m1, m8
    pmaddwd    m3, m8
%else
    pmaddwd    m0, [pw_1]
    pmaddwd    m2, [pw_1]
    pmaddwd    m1, [factor_p1_n1]
    pmaddwd    m3, [factor_p1_n1]
%endif

    paddd      m0, m4
    paddd      m2, m5
    paddd      m1, m6
    paddd      m3, m7

    psrad      m0, 1
    psrad      m2, 1
    psrad      m1, 1
    psrad      m3, 1

    packssdw   m0, m2
    packssdw   m1, m3

    mov      posq, ostrideq
    imul     posq, 2
    imul     posq, yq
    add      posq, xq

    movu    [outputq + posq], m0
    add      posq, ostrideq
    movu    [outputq + posq], m1

    add        yq, 1
    cmp        yq, heightq
    jl .looph

    mov      posq, lwidthq
    imul     posq, yq
    add      posq, xq
    movu       m0, [lowq + posq]
    sub      posq, lwidthq
    movu       m1, [lowq + posq]
    mova       m2, m0
    punpcklwd  m0, m1
    punpckhwd  m2, m1

%if ARCH_X86_64
    pmaddwd    m0, m13
    pmaddwd    m2, m13
%else
    pmaddwd    m0, [factor_p5_p4]
    pmaddwd    m2, [factor_p5_p4]
%endif

    pxor       m4, m4
    sub      posq, lwidthq
    movu       m1, [lowq + posq]
    mova       m3, m4
    punpcklwd  m4, m1
    punpckhwd  m3, m1

    psrad      m4, 16
    psrad      m3, 16

    psubd      m0, m4
    psubd      m2, m3

%if ARCH_X86_64
    paddd      m0, m11
    paddd      m2, m11
%else
    paddd      m0, [pd_4]
    paddd      m2, [pd_4]
%endif

    psrad      m0, 3
    psrad      m2, 3

    mov      posq, hwidthq
    imul     posq, yq
    add      posq, xq
    pxor       m4, m4
    movu       m1, [highq + posq]
    mova       m3, m4
    punpcklwd  m4, m1
    punpckhwd  m3, m1

    psrad      m4, 16
    psrad      m3, 16

    paddd      m0, m4
    paddd      m2, m3

    psrad      m0, 1
    psrad      m2, 1

    packssdw   m0, m2

    mov      posq, ostrideq
    imul     posq, 2
    imul     posq, yq
    add      posq, xq
    movu    [outputq + posq], m0

    mov      posq, lwidthq
    imul     posq, yq
    add      posq, xq
    movu       m0, [lowq + posq]
    sub      posq, lwidthq
    movu       m1, [lowq + posq]
    mova       m2, m0
    punpcklwd  m0, m1
    punpckhwd  m2, m1

%if ARCH_X86_64
    pmaddwd    m0, m12
    pmaddwd    m2, m12
%else
    pmaddwd    m0, [factor_p11_n4]
    pmaddwd    m2, [factor_p11_n4]
%endif

    pxor       m4, m4
    sub      posq, lwidthq
    movu       m1, [lowq + posq]
    mova       m3, m4
    punpcklwd  m4, m1
    punpckhwd  m3, m1

    psrad      m4, 16
    psrad      m3, 16

    paddd      m0, m4
    paddd      m2, m3

%if ARCH_X86_64
    paddd      m0, m11
    paddd      m2, m11
%else
    paddd      m0, [pd_4]
    paddd      m2, [pd_4]
%endif

    psrad      m0, 3
    psrad      m2, 3

    mov      posq, hwidthq
    imul     posq, yq
    add      posq, xq
    pxor       m4, m4
    movu       m1, [highq + posq]
    mova       m3, m4
    punpcklwd  m4, m1
    punpckhwd  m3, m1

    psrad      m4, 16
    psrad      m3, 16

    psubd      m0, m4
    psubd      m2, m3

    psrad      m0, 1
    psrad      m2, 1

    packssdw   m0, m2

    mov      posq, ostrideq
    imul     posq, 2
    imul     posq, yq
    add      posq, ostrideq
    add      posq, xq
    movu    [outputq + posq], m0

    add        xq, mmsize
    cmp        xq, widthq
    jl .loopw
    RET