aboutsummaryrefslogtreecommitdiffstats
path: root/libavcodec/arm/simple_idct_armv5te.S
blob: a8d03469abe01217931e0cbeca9d87f9bf0675e0 (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
/*
 * Simple IDCT
 *
 * Copyright (c) 2001 Michael Niedermayer <michaelni@gmx.at>
 * Copyright (c) 2006 Mans Rullgard <mans@mansr.com>
 *
 * 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/arm/asm.S"

#define W1  22725   /* cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5 */
#define W2  21407   /* cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5 */
#define W3  19266   /* cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5 */
#define W4  16383   /* cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5 */
#define W5  12873   /* cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5 */
#define W6  8867    /* cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5 */
#define W7  4520    /* cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5 */
#define ROW_SHIFT 11
#define COL_SHIFT 20

#define W13 (W1 | (W3 << 16))
#define W26 (W2 | (W6 << 16))
#define W57 (W5 | (W7 << 16))

function idct_row_armv5te
        str    lr, [sp, #-4]!

        ldrd   v1, v2, [a1, #8]
        ldrd   a3, a4, [a1]          /* a3 = row[1:0], a4 = row[3:2] */
        orrs   v1, v1, v2
        itt    eq
        cmpeq  v1, a4
        cmpeq  v1, a3, lsr #16
        beq    row_dc_only

        mov    v1, #(1<<(ROW_SHIFT-1))
        mov    ip, #16384
        sub    ip, ip, #1            /* ip = W4 */
        smlabb v1, ip, a3, v1        /* v1 = W4*row[0]+(1<<(RS-1)) */
        ldr    ip, =W26              /* ip = W2 | (W6 << 16) */
        smultb a2, ip, a4
        smulbb lr, ip, a4
        add    v2, v1, a2
        sub    v3, v1, a2
        sub    v4, v1, lr
        add    v1, v1, lr

        ldr    ip, =W13              /* ip = W1 | (W3 << 16) */
        ldr    lr, =W57              /* lr = W5 | (W7 << 16) */
        smulbt v5, ip, a3
        smultt v6, lr, a4
        smlatt v5, ip, a4, v5
        smultt a2, ip, a3
        smulbt v7, lr, a3
        sub    v6, v6, a2
        smulbt a2, ip, a4
        smultt fp, lr, a3
        sub    v7, v7, a2
        smulbt a2, lr, a4
        ldrd   a3, a4, [a1, #8]     /* a3=row[5:4] a4=row[7:6] */
        sub    fp, fp, a2

        orrs   a2, a3, a4
        beq    1f

        smlabt v5, lr, a3, v5
        smlabt v6, ip, a3, v6
        smlatt v5, lr, a4, v5
        smlabt v6, lr, a4, v6
        smlatt v7, lr, a3, v7
        smlatt fp, ip, a3, fp
        smulbt a2, ip, a4
        smlatt v7, ip, a4, v7
        sub    fp, fp, a2

        ldr    ip, =W26              /* ip = W2 | (W6 << 16) */
        mov    a2, #16384
        sub    a2, a2, #1            /* a2 =  W4 */
        smulbb a2, a2, a3            /* a2 =  W4*row[4] */
        smultb lr, ip, a4            /* lr =  W6*row[6] */
        add    v1, v1, a2            /* v1 += W4*row[4] */
        add    v1, v1, lr            /* v1 += W6*row[6] */
        add    v4, v4, a2            /* v4 += W4*row[4] */
        sub    v4, v4, lr            /* v4 -= W6*row[6] */
        smulbb lr, ip, a4            /* lr =  W2*row[6] */
        sub    v2, v2, a2            /* v2 -= W4*row[4] */
        sub    v2, v2, lr            /* v2 -= W2*row[6] */
        sub    v3, v3, a2            /* v3 -= W4*row[4] */
        add    v3, v3, lr            /* v3 += W2*row[6] */

1:      add    a2, v1, v5
        mov    a3, a2, lsr #11
        bic    a3, a3, #0x1f0000
        sub    a2, v2, v6
        mov    a2, a2, lsr #11
        add    a3, a3, a2, lsl #16
        add    a2, v3, v7
        mov    a4, a2, lsr #11
        bic    a4, a4, #0x1f0000
        add    a2, v4, fp
        mov    a2, a2, lsr #11
        add    a4, a4, a2, lsl #16
        strd   a3, a4, [a1]

        sub    a2, v4, fp
        mov    a3, a2, lsr #11
        bic    a3, a3, #0x1f0000
        sub    a2, v3, v7
        mov    a2, a2, lsr #11
        add    a3, a3, a2, lsl #16
        add    a2, v2, v6
        mov    a4, a2, lsr #11
        bic    a4, a4, #0x1f0000
        sub    a2, v1, v5
        mov    a2, a2, lsr #11
        add    a4, a4, a2, lsl #16
        strd   a3, a4, [a1, #8]

        ldr    pc, [sp], #4

row_dc_only:
        orr    a3, a3, a3, lsl #16
        bic    a3, a3, #0xe000
        mov    a3, a3, lsl #3
        mov    a4, a3
        strd   a3, a4, [a1]
        strd   a3, a4, [a1, #8]

        ldr    pc, [sp], #4
endfunc

        .macro idct_col
        ldr    a4, [a1]              /* a4 = col[1:0] */
        mov    ip, #16384
        sub    ip, ip, #1            /* ip = W4 */
        mov    v1, #((1<<(COL_SHIFT-1))/W4) /* this matches the C version */
        add    v2, v1, a4, asr #16
        rsb    v2, v2, v2, lsl #14
        mov    a4, a4, lsl #16
        add    v1, v1, a4, asr #16
        ldr    a4, [a1, #(16*4)]
        rsb    v1, v1, v1, lsl #14

        smulbb lr, ip, a4
        smulbt a3, ip, a4
        sub    v3, v1, lr
        sub    v5, v1, lr
        add    v7, v1, lr
        add    v1, v1, lr
        sub    v4, v2, a3
        sub    v6, v2, a3
        add    fp, v2, a3
        ldr    ip, =W26
        ldr    a4, [a1, #(16*2)]
        add    v2, v2, a3

        smulbb lr, ip, a4
        smultb a3, ip, a4
        add    v1, v1, lr
        sub    v7, v7, lr
        add    v3, v3, a3
        sub    v5, v5, a3
        smulbt lr, ip, a4
        smultt a3, ip, a4
        add    v2, v2, lr
        sub    fp, fp, lr
        add    v4, v4, a3
        ldr    a4, [a1, #(16*6)]
        sub    v6, v6, a3

        smultb lr, ip, a4
        smulbb a3, ip, a4
        add    v1, v1, lr
        sub    v7, v7, lr
        sub    v3, v3, a3
        add    v5, v5, a3
        smultt lr, ip, a4
        smulbt a3, ip, a4
        add    v2, v2, lr
        sub    fp, fp, lr
        sub    v4, v4, a3
        add    v6, v6, a3

        stmfd  sp!, {v1, v2, v3, v4, v5, v6, v7, fp}

        ldr    ip, =W13
        ldr    a4, [a1, #(16*1)]
        ldr    lr, =W57
        smulbb v1, ip, a4
        smultb v3, ip, a4
        smulbb v5, lr, a4
        smultb v7, lr, a4
        smulbt v2, ip, a4
        smultt v4, ip, a4
        smulbt v6, lr, a4
        smultt fp, lr, a4
        rsb    v4, v4, #0
        ldr    a4, [a1, #(16*3)]
        rsb    v3, v3, #0

        smlatb v1, ip, a4, v1
        smlatb v3, lr, a4, v3
        smulbb a3, ip, a4
        smulbb a2, lr, a4
        sub    v5, v5, a3
        sub    v7, v7, a2
        smlatt v2, ip, a4, v2
        smlatt v4, lr, a4, v4
        smulbt a3, ip, a4
        smulbt a2, lr, a4
        sub    v6, v6, a3
        ldr    a4, [a1, #(16*5)]
        sub    fp, fp, a2

        smlabb v1, lr, a4, v1
        smlabb v3, ip, a4, v3
        smlatb v5, lr, a4, v5
        smlatb v7, ip, a4, v7
        smlabt v2, lr, a4, v2
        smlabt v4, ip, a4, v4
        smlatt v6, lr, a4, v6
        ldr    a3, [a1, #(16*7)]
        smlatt fp, ip, a4, fp

        smlatb v1, lr, a3, v1
        smlabb v3, lr, a3, v3
        smlatb v5, ip, a3, v5
        smulbb a4, ip, a3
        smlatt v2, lr, a3, v2
        sub    v7, v7, a4
        smlabt v4, lr, a3, v4
        smulbt a4, ip, a3
        smlatt v6, ip, a3, v6
        sub    fp, fp, a4
        .endm

function idct_col_armv5te
        str    lr, [sp, #-4]!

        idct_col

        ldmfd  sp!, {a3, a4}
        adds   a2, a3, v1
        mov    a2, a2, lsr #20
        it     mi
        orrmi  a2, a2, #0xf000
        add    ip, a4, v2
        mov    ip, ip, asr #20
        orr    a2, a2, ip, lsl #16
        str    a2, [a1]
        subs   a3, a3, v1
        mov    a2, a3, lsr #20
        it     mi
        orrmi  a2, a2, #0xf000
        sub    a4, a4, v2
        mov    a4, a4, asr #20
        orr    a2, a2, a4, lsl #16
        ldmfd  sp!, {a3, a4}
        str    a2, [a1, #(16*7)]

        subs   a2, a3, v3
        mov    a2, a2, lsr #20
        it     mi
        orrmi  a2, a2, #0xf000
        sub    ip, a4, v4
        mov    ip, ip, asr #20
        orr    a2, a2, ip, lsl #16
        str    a2, [a1, #(16*1)]
        adds   a3, a3, v3
        mov    a2, a3, lsr #20
        it     mi
        orrmi  a2, a2, #0xf000
        add    a4, a4, v4
        mov    a4, a4, asr #20
        orr    a2, a2, a4, lsl #16
        ldmfd  sp!, {a3, a4}
        str    a2, [a1, #(16*6)]

        adds   a2, a3, v5
        mov    a2, a2, lsr #20
        it     mi
        orrmi  a2, a2, #0xf000
        add    ip, a4, v6
        mov    ip, ip, asr #20
        orr    a2, a2, ip, lsl #16
        str    a2, [a1, #(16*2)]
        subs   a3, a3, v5
        mov    a2, a3, lsr #20
        it     mi
        orrmi  a2, a2, #0xf000
        sub    a4, a4, v6
        mov    a4, a4, asr #20
        orr    a2, a2, a4, lsl #16
        ldmfd  sp!, {a3, a4}
        str    a2, [a1, #(16*5)]

        adds   a2, a3, v7
        mov    a2, a2, lsr #20
        it     mi
        orrmi  a2, a2, #0xf000
        add    ip, a4, fp
        mov    ip, ip, asr #20
        orr    a2, a2, ip, lsl #16
        str    a2, [a1, #(16*3)]
        subs   a3, a3, v7
        mov    a2, a3, lsr #20
        it     mi
        orrmi  a2, a2, #0xf000
        sub    a4, a4, fp
        mov    a4, a4, asr #20
        orr    a2, a2, a4, lsl #16
        str    a2, [a1, #(16*4)]

        ldr    pc, [sp], #4
endfunc

.macro  clip   dst, src:vararg
        movs   \dst, \src
        it     mi
        movmi  \dst, #0
        cmp    \dst, #255
        it     gt
        movgt  \dst, #255
.endm

.macro  aclip  dst, src:vararg
        adds   \dst, \src
        it     mi
        movmi  \dst, #0
        cmp    \dst, #255
        it     gt
        movgt  \dst, #255
.endm

function idct_col_put_armv5te
        str    lr, [sp, #-4]!

        idct_col

        ldmfd  sp!, {a3, a4}
        ldr    lr, [sp, #32]
        add    a2, a3, v1
        clip   a2, a2, asr #20
        add    ip, a4, v2
        clip   ip, ip, asr #20
        orr    a2, a2, ip, lsl #8
        sub    a3, a3, v1
        clip   a3, a3, asr #20
        sub    a4, a4, v2
        clip   a4, a4, asr #20
        ldr    v1, [sp, #28]
        strh   a2, [v1]
        add    a2, v1, #2
        str    a2, [sp, #28]
        orr    a2, a3, a4, lsl #8
        rsb    v2, lr, lr, lsl #3
        ldmfd  sp!, {a3, a4}
        strh_pre a2, v2, v1

        sub    a2, a3, v3
        clip   a2, a2, asr #20
        sub    ip, a4, v4
        clip   ip, ip, asr #20
        orr    a2, a2, ip, lsl #8
        strh_pre a2, v1, lr
        add    a3, a3, v3
        clip   a2, a3, asr #20
        add    a4, a4, v4
        clip   a4, a4, asr #20
        orr    a2, a2, a4, lsl #8
        ldmfd  sp!, {a3, a4}
        strh_dpre a2, v2, lr

        add    a2, a3, v5
        clip   a2, a2, asr #20
        add    ip, a4, v6
        clip   ip, ip, asr #20
        orr    a2, a2, ip, lsl #8
        strh_pre a2, v1, lr
        sub    a3, a3, v5
        clip   a2, a3, asr #20
        sub    a4, a4, v6
        clip   a4, a4, asr #20
        orr    a2, a2, a4, lsl #8
        ldmfd  sp!, {a3, a4}
        strh_dpre a2, v2, lr

        add    a2, a3, v7
        clip   a2, a2, asr #20
        add    ip, a4, fp
        clip   ip, ip, asr #20
        orr    a2, a2, ip, lsl #8
        strh   a2, [v1, lr]
        sub    a3, a3, v7
        clip   a2, a3, asr #20
        sub    a4, a4, fp
        clip   a4, a4, asr #20
        orr    a2, a2, a4, lsl #8
        strh_dpre a2, v2, lr

        ldr    pc, [sp], #4
endfunc

function idct_col_add_armv5te
        str    lr, [sp, #-4]!

        idct_col

        ldr    lr, [sp, #36]

        ldmfd  sp!, {a3, a4}
        ldrh   ip, [lr]
        add    a2, a3, v1
        sub    a3, a3, v1
        and    v1, ip, #255
        aclip  a2, v1, a2, asr #20
        add    v1, a4, v2
        mov    v1, v1, asr #20
        aclip  v1, v1, ip, lsr #8
        orr    a2, a2, v1, lsl #8
        ldr    v1, [sp, #32]
        sub    a4, a4, v2
        rsb    v2, v1, v1, lsl #3
        ldrh_pre ip, v2, lr
        strh   a2, [lr]
        and    a2, ip, #255
        aclip  a3, a2, a3, asr #20
        mov    a4, a4, asr #20
        aclip  a4, a4, ip, lsr #8
        add    a2, lr, #2
        str    a2, [sp, #28]
        orr    a2, a3, a4, lsl #8
        strh   a2, [v2]

        ldmfd  sp!, {a3, a4}
        ldrh_pre ip, lr, v1
        sub    a2, a3, v3
        add    a3, a3, v3
        and    v3, ip, #255
        aclip  a2, v3, a2, asr #20
        sub    v3, a4, v4
        mov    v3, v3, asr #20
        aclip  v3, v3, ip, lsr #8
        orr    a2, a2, v3, lsl #8
        add    a4, a4, v4
        ldrh_dpre ip, v2, v1
        strh   a2, [lr]
        and    a2, ip, #255
        aclip  a3, a2, a3, asr #20
        mov    a4, a4, asr #20
        aclip  a4, a4, ip, lsr #8
        orr    a2, a3, a4, lsl #8
        strh   a2, [v2]

        ldmfd  sp!, {a3, a4}
        ldrh_pre ip, lr, v1
        add    a2, a3, v5
        sub    a3, a3, v5
        and    v3, ip, #255
        aclip  a2, v3, a2, asr #20
        add    v3, a4, v6
        mov    v3, v3, asr #20
        aclip  v3, v3, ip, lsr #8
        orr    a2, a2, v3, lsl #8
        sub    a4, a4, v6
        ldrh_dpre ip, v2, v1
        strh   a2, [lr]
        and    a2, ip, #255
        aclip  a3, a2, a3, asr #20
        mov    a4, a4, asr #20
        aclip  a4, a4, ip, lsr #8
        orr    a2, a3, a4, lsl #8
        strh   a2, [v2]

        ldmfd  sp!, {a3, a4}
        ldrh_pre ip, lr, v1
        add    a2, a3, v7
        sub    a3, a3, v7
        and    v3, ip, #255
        aclip  a2, v3, a2, asr #20
        add    v3, a4, fp
        mov    v3, v3, asr #20
        aclip  v3, v3, ip, lsr #8
        orr    a2, a2, v3, lsl #8
        sub    a4, a4, fp
        ldrh_dpre ip, v2, v1
        strh   a2, [lr]
        and    a2, ip, #255
        aclip  a3, a2, a3, asr #20
        mov    a4, a4, asr #20
        aclip  a4, a4, ip, lsr #8
        orr    a2, a3, a4, lsl #8
        strh   a2, [v2]

        ldr    pc, [sp], #4
endfunc

function ff_simple_idct_armv5te, export=1
        stmfd  sp!, {v1, v2, v3, v4, v5, v6, v7, fp, lr}

        bl     idct_row_armv5te
        add    a1, a1, #16
        bl     idct_row_armv5te
        add    a1, a1, #16
        bl     idct_row_armv5te
        add    a1, a1, #16
        bl     idct_row_armv5te
        add    a1, a1, #16
        bl     idct_row_armv5te
        add    a1, a1, #16
        bl     idct_row_armv5te
        add    a1, a1, #16
        bl     idct_row_armv5te
        add    a1, a1, #16
        bl     idct_row_armv5te

        sub    a1, a1, #(16*7)

        bl     idct_col_armv5te
        add    a1, a1, #4
        bl     idct_col_armv5te
        add    a1, a1, #4
        bl     idct_col_armv5te
        add    a1, a1, #4
        bl     idct_col_armv5te

        ldmfd  sp!, {v1, v2, v3, v4, v5, v6, v7, fp, pc}
endfunc

function ff_simple_idct_add_armv5te, export=1
        stmfd  sp!, {a1, a2, v1, v2, v3, v4, v5, v6, v7, fp, lr}

        mov    a1, a3

        bl     idct_row_armv5te
        add    a1, a1, #16
        bl     idct_row_armv5te
        add    a1, a1, #16
        bl     idct_row_armv5te
        add    a1, a1, #16
        bl     idct_row_armv5te
        add    a1, a1, #16
        bl     idct_row_armv5te
        add    a1, a1, #16
        bl     idct_row_armv5te
        add    a1, a1, #16
        bl     idct_row_armv5te
        add    a1, a1, #16
        bl     idct_row_armv5te

        sub    a1, a1, #(16*7)

        bl     idct_col_add_armv5te
        add    a1, a1, #4
        bl     idct_col_add_armv5te
        add    a1, a1, #4
        bl     idct_col_add_armv5te
        add    a1, a1, #4
        bl     idct_col_add_armv5te

        add    sp, sp, #8
        ldmfd  sp!, {v1, v2, v3, v4, v5, v6, v7, fp, pc}
endfunc

function ff_simple_idct_put_armv5te, export=1
        stmfd  sp!, {a1, a2, v1, v2, v3, v4, v5, v6, v7, fp, lr}

        mov    a1, a3

        bl     idct_row_armv5te
        add    a1, a1, #16
        bl     idct_row_armv5te
        add    a1, a1, #16
        bl     idct_row_armv5te
        add    a1, a1, #16
        bl     idct_row_armv5te
        add    a1, a1, #16
        bl     idct_row_armv5te
        add    a1, a1, #16
        bl     idct_row_armv5te
        add    a1, a1, #16
        bl     idct_row_armv5te
        add    a1, a1, #16
        bl     idct_row_armv5te

        sub    a1, a1, #(16*7)

        bl     idct_col_put_armv5te
        add    a1, a1, #4
        bl     idct_col_put_armv5te
        add    a1, a1, #4
        bl     idct_col_put_armv5te
        add    a1, a1, #4
        bl     idct_col_put_armv5te

        add    sp, sp, #8
        ldmfd  sp!, {v1, v2, v3, v4, v5, v6, v7, fp, pc}
endfunc