summaryrefslogtreecommitdiff
path: root/libavcodec/arm/dsputil_arm.S
blob: 1247b0fa84c39a31ae4fb4224fe592a78a454946 (plain)
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
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
@
@ ARMv4 optimized DSP utils
@ Copyright (c) 2004 AGAWA Koji <i (AT) atty (DOT) jp>
@
@ 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 "config.h"
#include "asm.S"

        preserve8

#if HAVE_ARMV5TE
function ff_prefetch_arm, export=1
        subs            r2,  r2,  #1
        pld             [r0]
        add             r0,  r0,  r1
        bne             ff_prefetch_arm
        bx              lr
endfunc
#else
#define pld @
#endif

.macro  ALIGN_QWORD_D shift, Rd0, Rd1, Rd2, Rd3, Rn0, Rn1, Rn2, Rn3, Rn4
        mov             \Rd0, \Rn0, lsr #(\shift * 8)
        mov             \Rd1, \Rn1, lsr #(\shift * 8)
        mov             \Rd2, \Rn2, lsr #(\shift * 8)
        mov             \Rd3, \Rn3, lsr #(\shift * 8)
        orr             \Rd0, \Rd0, \Rn1, lsl #(32 - \shift * 8)
        orr             \Rd1, \Rd1, \Rn2, lsl #(32 - \shift * 8)
        orr             \Rd2, \Rd2, \Rn3, lsl #(32 - \shift * 8)
        orr             \Rd3, \Rd3, \Rn4, lsl #(32 - \shift * 8)
.endm
.macro  ALIGN_DWORD shift, R0, R1, R2
        mov             \R0, \R0, lsr #(\shift * 8)
        orr             \R0, \R0, \R1, lsl #(32 - \shift * 8)
        mov             \R1, \R1, lsr #(\shift * 8)
        orr             \R1, \R1, \R2, lsl #(32 - \shift * 8)
.endm
.macro  ALIGN_DWORD_D shift, Rdst0, Rdst1, Rsrc0, Rsrc1, Rsrc2
        mov             \Rdst0, \Rsrc0, lsr #(\shift * 8)
        mov             \Rdst1, \Rsrc1, lsr #(\shift * 8)
        orr             \Rdst0, \Rdst0, \Rsrc1, lsl #(32 - (\shift * 8))
        orr             \Rdst1, \Rdst1, \Rsrc2, lsl #(32 - (\shift * 8))
.endm

.macro  RND_AVG32 Rd0, Rd1, Rn0, Rn1, Rm0, Rm1, Rmask
        @ Rd = (Rn | Rm) - (((Rn ^ Rm) & ~0x01010101) >> 1)
        @ Rmask = 0xFEFEFEFE
        @ Rn = destroy
        eor             \Rd0, \Rn0, \Rm0
        eor             \Rd1, \Rn1, \Rm1
        orr             \Rn0, \Rn0, \Rm0
        orr             \Rn1, \Rn1, \Rm1
        and             \Rd0, \Rd0, \Rmask
        and             \Rd1, \Rd1, \Rmask
        sub             \Rd0, \Rn0, \Rd0, lsr #1
        sub             \Rd1, \Rn1, \Rd1, lsr #1
.endm

.macro  NO_RND_AVG32 Rd0, Rd1, Rn0, Rn1, Rm0, Rm1, Rmask
        @ Rd = (Rn & Rm) - (((Rn ^ Rm) & ~0x01010101) >> 1)
        @ Rmask = 0xFEFEFEFE
        @ Rn = destroy
        eor             \Rd0, \Rn0, \Rm0
        eor             \Rd1, \Rn1, \Rm1
        and             \Rn0, \Rn0, \Rm0
        and             \Rn1, \Rn1, \Rm1
        and             \Rd0, \Rd0, \Rmask
        and             \Rd1, \Rd1, \Rmask
        add             \Rd0, \Rn0, \Rd0, lsr #1
        add             \Rd1, \Rn1, \Rd1, lsr #1
.endm

.macro  JMP_ALIGN tmp, reg
        ands            \tmp, \reg, #3
        bic             \reg, \reg, #3
        beq             1f
        subs            \tmp, \tmp, #1
        beq             2f
        subs            \tmp, \tmp, #1
        beq             3f
        b    4f
.endm

@ ----------------------------------------------------------------
        .align 5
function ff_put_pixels16_arm, export=1
        @ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h)
        @ block = word aligned, pixles = unaligned
        pld             [r1]
        push            {r4-r11, lr}
        JMP_ALIGN       r5,  r1
1:
        ldm             r1,  {r4-r7}
        add             r1,  r1,  r2
        stm             r0,  {r4-r7}
        pld             [r1]
        subs            r3,  r3,  #1
        add             r0,  r0,  r2
        bne             1b
        pop             {r4-r11, pc}
        .align 5
2:
        ldm             r1,  {r4-r8}
        add             r1,  r1,  r2
        ALIGN_QWORD_D   1,   r9,  r10, r11, r12, r4,  r5,  r6,  r7,  r8
        pld             [r1]
        subs            r3,  r3,  #1
        stm             r0,  {r9-r12}
        add             r0,  r0,  r2
        bne             2b
        pop             {r4-r11, pc}
        .align 5
3:
        ldm             r1,  {r4-r8}
        add             r1,  r1,  r2
        ALIGN_QWORD_D   2,   r9,  r10, r11, r12, r4,  r5,  r6,  r7,  r8
        pld             [r1]
        subs            r3,  r3,  #1
        stm             r0,  {r9-r12}
        add             r0,  r0,  r2
        bne             3b
        pop             {r4-r11, pc}
        .align 5
4:
        ldm             r1,  {r4-r8}
        add             r1,  r1,  r2
        ALIGN_QWORD_D   3,   r9,  r10, r11, r12, r4,  r5,  r6,  r7,  r8
        pld             [r1]
        subs            r3,  r3,  #1
        stm             r0,  {r9-r12}
        add             r0,  r0,  r2
        bne             4b
        pop             {r4-r11,pc}
endfunc

@ ----------------------------------------------------------------
        .align 5
function ff_put_pixels8_arm, export=1
        @ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h)
        @ block = word aligned, pixles = unaligned
        pld             [r1]
        push            {r4-r5,lr}
        JMP_ALIGN       r5,  r1
1:
        ldm             r1,  {r4-r5}
        add             r1,  r1,  r2
        subs            r3,  r3,  #1
        pld             [r1]
        stm             r0,  {r4-r5}
        add             r0,  r0,  r2
        bne             1b
        pop             {r4-r5,pc}
        .align 5
2:
        ldm             r1,  {r4-r5, r12}
        add             r1,  r1,  r2
        ALIGN_DWORD     1,   r4,  r5,  r12
        pld             [r1]
        subs            r3,  r3,  #1
        stm             r0,  {r4-r5}
        add             r0,  r0,  r2
        bne             2b
        pop             {r4-r5,pc}
        .align 5
3:
        ldm             r1,  {r4-r5, r12}
        add             r1,  r1,  r2
        ALIGN_DWORD     2,   r4,  r5,  r12
        pld             [r1]
        subs            r3,  r3,  #1
        stm             r0,  {r4-r5}
        add             r0,  r0,  r2
        bne             3b
        pop             {r4-r5,pc}
        .align 5
4:
        ldm             r1,  {r4-r5, r12}
        add             r1,  r1,  r2
        ALIGN_DWORD     3,   r4,  r5,  r12
        pld             [r1]
        subs            r3,  r3,  #1
        stm             r0,  {r4-r5}
        add             r0,  r0,  r2
        bne             4b
        pop             {r4-r5,pc}
endfunc

@ ----------------------------------------------------------------
        .align 5
function ff_put_pixels8_x2_arm, export=1
        @ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h)
        @ block = word aligned, pixles = unaligned
        pld             [r1]
        push            {r4-r10,lr}
        ldr             r12, =0xfefefefe
        JMP_ALIGN       r5,  r1
1:
        ldm             r1,  {r4-r5, r10}
        add             r1,  r1,  r2
        ALIGN_DWORD_D   1,   r6,  r7,  r4,  r5,  r10
        pld             [r1]
        RND_AVG32       r8,  r9,  r4,  r5,  r6,  r7,  r12
        subs            r3,  r3,  #1
        stm             r0,  {r8-r9}
        add             r0,  r0,  r2
        bne             1b
        pop             {r4-r10,pc}
        .align 5
2:
        ldm             r1,  {r4-r5, r10}
        add             r1,  r1,  r2
        ALIGN_DWORD_D   1,   r6,  r7,  r4,  r5,  r10
        ALIGN_DWORD_D   2,   r8,  r9,  r4,  r5,  r10
        pld             [r1]
        RND_AVG32       r4,  r5,  r6,  r7,  r8,  r9,  r12
        subs            r3,  r3,  #1
        stm             r0,  {r4-r5}
        add             r0,  r0,  r2
        bne             2b
        pop             {r4-r10,pc}
        .align 5
3:
        ldm             r1,  {r4-r5, r10}
        add             r1,  r1,  r2
        ALIGN_DWORD_D   2,   r6,  r7,  r4,  r5,  r10
        ALIGN_DWORD_D   3,   r8,  r9,  r4,  r5,  r10
        pld             [r1]
        RND_AVG32       r4,  r5,  r6,  r7,  r8,  r9,  r12
        subs            r3,  r3,  #1
        stm             r0,  {r4-r5}
        add             r0,  r0,  r2
        bne             3b
        pop             {r4-r10,pc}
        .align 5
4:
        ldm             r1,  {r4-r5, r10}
        add             r1,  r1,  r2
        ALIGN_DWORD_D   3,   r6,  r7,  r4,  r5,  r10
        pld             [r1]
        RND_AVG32       r8,  r9,  r6,  r7,  r5,  r10, r12
        subs            r3,  r3,  #1
        stm             r0,  {r8-r9}
        add             r0,  r0,  r2
        bne             4b
        pop             {r4-r10,pc}
endfunc

        .align 5
function ff_put_no_rnd_pixels8_x2_arm, export=1
        @ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h)
        @ block = word aligned, pixles = unaligned
        pld             [r1]
        push            {r4-r10,lr}
        ldr             r12, =0xfefefefe
        JMP_ALIGN       r5,  r1
1:
        ldm             r1,  {r4-r5, r10}
        add             r1,  r1,  r2
        ALIGN_DWORD_D   1,   r6,  r7,  r4,  r5,  r10
        pld             [r1]
        NO_RND_AVG32    r8,  r9,  r4,  r5,  r6,  r7,  r12
        subs            r3,  r3,  #1
        stm             r0,  {r8-r9}
        add             r0,  r0,  r2
        bne             1b
        pop             {r4-r10,pc}
        .align 5
2:
        ldm             r1,  {r4-r5, r10}
        add             r1,  r1,  r2
        ALIGN_DWORD_D   1,   r6,  r7,  r4,  r5,  r10
        ALIGN_DWORD_D   2,   r8,  r9,  r4,  r5,  r10
        pld             [r1]
        NO_RND_AVG32    r4,  r5,  r6,  r7,  r8,  r9,  r12
        subs            r3,  r3,  #1
        stm             r0,  {r4-r5}
        add             r0,  r0,  r2
        bne             2b
        pop             {r4-r10,pc}
        .align 5
3:
        ldm             r1,  {r4-r5, r10}
        add             r1,  r1,  r2
        ALIGN_DWORD_D   2,   r6,  r7,  r4,  r5,  r10
        ALIGN_DWORD_D   3,   r8,  r9,  r4,  r5,  r10
        pld             [r1]
        NO_RND_AVG32    r4,  r5,  r6,  r7,  r8,  r9,  r12
        subs            r3,  r3,  #1
        stm             r0,  {r4-r5}
        add             r0,  r0,  r2
        bne             3b
        pop             {r4-r10,pc}
        .align 5
4:
        ldm             r1,  {r4-r5, r10}
        add             r1,  r1,  r2
        ALIGN_DWORD_D   3,   r6,  r7,  r4,  r5,  r10
        pld             [r1]
        NO_RND_AVG32    r8,  r9,  r6,  r7,  r5,  r10, r12
        subs            r3,  r3,  #1
        stm             r0,  {r8-r9}
        add             r0,  r0,  r2
        bne             4b
        pop             {r4-r10,pc}
endfunc


@ ----------------------------------------------------------------
        .align 5
function ff_put_pixels8_y2_arm, export=1
        @ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h)
        @ block = word aligned, pixles = unaligned
        pld             [r1]
        push            {r4-r11,lr}
        mov             r3,  r3,  lsr #1
        ldr             r12, =0xfefefefe
        JMP_ALIGN       r5,  r1
1:
        ldm             r1,  {r4-r5}
        add             r1,  r1,  r2
6:      ldm             r1,  {r6-r7}
        add             r1,  r1,  r2
        pld             [r1]
        RND_AVG32       r8,  r9,  r4,  r5,  r6,  r7,  r12
        ldm             r1,  {r4-r5}
        add             r1,  r1,  r2
        stm             r0,  {r8-r9}
        add             r0,  r0,  r2
        pld             [r1]
        RND_AVG32       r8,  r9,  r6,  r7,  r4,  r5,  r12
        subs            r3,  r3,  #1
        stm             r0,  {r8-r9}
        add             r0,  r0,  r2
        bne             6b
        pop             {r4-r11,pc}
        .align 5
2:
        ldm             r1,  {r4-r6}
        add             r1,  r1,  r2
        pld             [r1]
        ALIGN_DWORD     1,   r4,  r5,  r6
6:      ldm             r1,  {r7-r9}
        add             r1,  r1,  r2
        pld             [r1]
        ALIGN_DWORD     1,   r7,  r8,  r9
        RND_AVG32       r10, r11, r4,  r5,  r7,  r8,  r12
        stm             r0,  {r10-r11}
        add             r0,  r0,  r2
        ldm             r1,  {r4-r6}
        add             r1,  r1,  r2
        pld             [r1]
        ALIGN_DWORD     1,   r4,  r5,  r6
        subs            r3,  r3,  #1
        RND_AVG32       r10, r11, r7,  r8,  r4,  r5,  r12
        stm             r0,  {r10-r11}
        add             r0,  r0,  r2
        bne             6b
        pop             {r4-r11,pc}
        .align 5
3:
        ldm             r1,  {r4-r6}
        add             r1,  r1,  r2
        pld             [r1]
        ALIGN_DWORD     2,   r4,  r5,  r6
6:      ldm             r1,  {r7-r9}
        add             r1,  r1,  r2
        pld             [r1]
        ALIGN_DWORD     2,   r7,  r8,  r9
        RND_AVG32       r10, r11, r4,  r5,  r7,  r8,  r12
        stm             r0,  {r10-r11}
        add             r0,  r0,  r2
        ldm             r1,  {r4-r6}
        add             r1,  r1,  r2
        pld             [r1]
        ALIGN_DWORD     2,   r4,  r5,  r6
        subs            r3,  r3,  #1
        RND_AVG32       r10, r11, r7,  r8,  r4,  r5,  r12
        stm             r0,  {r10-r11}
        add             r0,  r0,  r2
        bne             6b
        pop             {r4-r11,pc}
        .align 5
4:
        ldm             r1,  {r4-r6}
        add             r1,  r1,  r2
        pld             [r1]
        ALIGN_DWORD     3,   r4,  r5,  r6
6:      ldm             r1,  {r7-r9}
        add             r1,  r1,  r2
        pld             [r1]
        ALIGN_DWORD     3,   r7,  r8,  r9
        RND_AVG32       r10, r11, r4,  r5,  r7,  r8,  r12
        stm             r0,  {r10-r11}
        add             r0,  r0,  r2
        ldm             r1,  {r4-r6}
        add             r1,  r1,  r2
        pld             [r1]
        ALIGN_DWORD     3,   r4,  r5,  r6
        subs            r3,  r3,  #1
        RND_AVG32       r10, r11, r7,  r8,  r4,  r5,  r12
        stm             r0,  {r10-r11}
        add             r0,  r0,  r2
        bne             6b
        pop             {r4-r11,pc}
endfunc

        .align 5
function ff_put_no_rnd_pixels8_y2_arm, export=1
        @ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h)
        @ block = word aligned, pixles = unaligned
        pld             [r1]
        push            {r4-r11,lr}
        mov             r3,  r3,  lsr #1
        ldr             r12, =0xfefefefe
        JMP_ALIGN       r5,  r1
1:
        ldm             r1,  {r4-r5}
        add             r1,  r1,  r2
6:      ldm             r1,  {r6-r7}
        add             r1,  r1,  r2
        pld             [r1]
        NO_RND_AVG32    r8,  r9,  r4,  r5,  r6,  r7,  r12
        ldm             r1,  {r4-r5}
        add             r1,  r1,  r2
        stm             r0,  {r8-r9}
        add             r0,  r0,  r2
        pld             [r1]
        NO_RND_AVG32    r8,  r9,  r6,  r7,  r4,  r5,  r12
        subs            r3,  r3,  #1
        stm             r0,  {r8-r9}
        add             r0,  r0,  r2
        bne             6b
        pop             {r4-r11,pc}
        .align 5
2:
        ldm             r1,  {r4-r6}
        add             r1,  r1,  r2
        pld             [r1]
        ALIGN_DWORD     1,   r4,  r5,  r6
6:      ldm             r1,  {r7-r9}
        add             r1,  r1,  r2
        pld             [r1]
        ALIGN_DWORD     1,   r7,  r8,  r9
        NO_RND_AVG32    r10, r11, r4,  r5,  r7,  r8,  r12
        stm             r0,  {r10-r11}
        add             r0,  r0,  r2
        ldm             r1,  {r4-r6}
        add             r1,  r1,  r2
        pld             [r1]
        ALIGN_DWORD     1,   r4,  r5,  r6
        subs            r3,  r3,  #1
        NO_RND_AVG32    r10, r11, r7,  r8,  r4,  r5,  r12
        stm             r0,  {r10-r11}
        add             r0,  r0,  r2
        bne             6b
        pop             {r4-r11,pc}
        .align 5
3:
        ldm             r1,  {r4-r6}
        add             r1,  r1,  r2
        pld             [r1]
        ALIGN_DWORD     2,   r4,  r5,  r6
6:      ldm             r1,  {r7-r9}
        add             r1,  r1,  r2
        pld             [r1]
        ALIGN_DWORD     2,   r7,  r8,  r9
        NO_RND_AVG32    r10, r11, r4,  r5,  r7,  r8,  r12
        stm             r0,  {r10-r11}
        add             r0,  r0,  r2
        ldm             r1,  {r4-r6}
        add             r1,  r1,  r2
        pld             [r1]
        ALIGN_DWORD     2,   r4,  r5,  r6
        subs            r3,  r3,  #1
        NO_RND_AVG32    r10, r11, r7,  r8,  r4,  r5,  r12
        stm             r0,  {r10-r11}
        add             r0,  r0,  r2
        bne             6b
        pop             {r4-r11,pc}
        .align 5
4:
        ldm             r1,  {r4-r6}
        add             r1,  r1,  r2
        pld             [r1]
        ALIGN_DWORD     3,   r4,  r5,  r6
6:      ldm             r1,  {r7-r9}
        add             r1,  r1,  r2
        pld             [r1]
        ALIGN_DWORD     3,   r7,  r8,  r9
        NO_RND_AVG32    r10, r11, r4,  r5,  r7,  r8,  r12
        stm             r0,  {r10-r11}
        add             r0,  r0,  r2
        ldm             r1,  {r4-r6}
        add             r1,  r1,  r2
        pld             [r1]
        ALIGN_DWORD     3,   r4,  r5,  r6
        subs            r3,  r3,  #1
        NO_RND_AVG32    r10, r11, r7,  r8,  r4,  r5,  r12
        stm             r0,  {r10-r11}
        add             r0,  r0,  r2
        bne             6b
        pop             {r4-r11,pc}
endfunc

        .ltorg

@ ----------------------------------------------------------------
.macro  RND_XY2_IT align, rnd
        @ l1=  (a & 0x03030303) + (b & 0x03030303) ?(+ 0x02020202)
        @ h1= ((a & 0xFCFCFCFCUL) >> 2) + ((b & 0xFCFCFCFCUL) >> 2)
.if \align == 0
        ldm             r1,  {r6-r8}
.elseif \align == 3
        ldm             r1,  {r5-r7}
.else
        ldm             r1,  {r8-r10}
.endif
        add             r1,  r1,  r2
        pld             [r1]
.if \align == 0
        ALIGN_DWORD_D   1,   r4,  r5,  r6,  r7,  r8
.elseif \align == 1
        ALIGN_DWORD_D   1,   r4,  r5,  r8,  r9,  r10
        ALIGN_DWORD_D   2,   r6,  r7,  r8,  r9,  r10
.elseif \align == 2
        ALIGN_DWORD_D   2,   r4,  r5,  r8,  r9,  r10
        ALIGN_DWORD_D   3,   r6,  r7,  r8,  r9,  r10
.elseif \align == 3
        ALIGN_DWORD_D   3,   r4,  r5,  r5,  r6,  r7
.endif
        ldr             r14, =0x03030303
        tst             r3,  #1
        and             r8,  r4,  r14
        and             r9,  r5,  r14
        and             r10, r6,  r14
        and             r11, r7,  r14
        it              eq
        andeq           r14, r14, r14, \rnd #1
        add             r8,  r8,  r10
        add             r9,  r9,  r11
        ldr             r12, =0xfcfcfcfc >> 2
        itt             eq
        addeq           r8,  r8,  r14
        addeq           r9,  r9,  r14
        and             r4,  r12, r4,  lsr #2
        and             r5,  r12, r5,  lsr #2
        and             r6,  r12, r6,  lsr #2
        and             r7,  r12, r7,  lsr #2
        add             r10, r4,  r6
        add             r11, r5,  r7
        subs            r3,  r3,  #1
.endm

.macro RND_XY2_EXPAND align, rnd
        RND_XY2_IT      \align, \rnd
6:      push            {r8-r11}
        RND_XY2_IT      \align, \rnd
        pop             {r4-r7}
        add             r4,  r4,  r8
        add             r5,  r5,  r9
        ldr             r14, =0x0f0f0f0f
        add             r6,  r6,  r10
        add             r7,  r7,  r11
        and             r4,  r14, r4,  lsr #2
        and             r5,  r14, r5,  lsr #2
        add             r4,  r4,  r6
        add             r5,  r5,  r7
        stm             r0,  {r4-r5}
        add             r0,  r0,  r2
        bge             6b
        pop             {r4-r11,pc}
.endm

        .align 5
function ff_put_pixels8_xy2_arm, export=1
        @ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h)
        @ block = word aligned, pixles = unaligned
        pld             [r1]
        push            {r4-r11,lr} @ R14 is also called LR
        JMP_ALIGN       r5,  r1
1:      RND_XY2_EXPAND  0, lsl
        .align 5
2:      RND_XY2_EXPAND  1, lsl
        .align 5
3:      RND_XY2_EXPAND  2, lsl
        .align 5
4:      RND_XY2_EXPAND  3, lsl
endfunc

        .align 5
function ff_put_no_rnd_pixels8_xy2_arm, export=1
        @ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h)
        @ block = word aligned, pixles = unaligned
        pld             [r1]
        push            {r4-r11,lr}
        JMP_ALIGN       r5,  r1
1:      RND_XY2_EXPAND  0, lsr
        .align 5
2:      RND_XY2_EXPAND  1, lsr
        .align 5
3:      RND_XY2_EXPAND  2, lsr
        .align 5
4:      RND_XY2_EXPAND  3, lsr
endfunc

        .align 5
@ void ff_add_pixels_clamped_arm(int16_t *block, uint8_t *dest, int stride)
function ff_add_pixels_clamped_arm, export=1
        push            {r4-r10}
        mov             r10, #8
1:
        ldr             r4,  [r1]               /* load dest */
        /* block[0] and block[1]*/
        ldrsh           r5,  [r0]
        ldrsh           r7,  [r0, #2]
        and             r6,  r4,  #0xFF
        and             r8,  r4,  #0xFF00
        add             r6,  r5,  r6
        add             r8,  r7,  r8,  lsr #8
        mvn             r5,  r5
        mvn             r7,  r7
        tst             r6,  #0x100
        it              ne
        movne           r6,  r5,  lsr #24
        tst             r8,  #0x100
        it              ne
        movne           r8,  r7,  lsr #24
        mov             r9,  r6
        ldrsh           r5,  [r0, #4]           /* moved form [A] */
        orr             r9,  r9,  r8,  lsl #8
        /* block[2] and block[3] */
        /* [A] */
        ldrsh           r7,  [r0, #6]
        and             r6,  r4,  #0xFF0000
        and             r8,  r4,  #0xFF000000
        add             r6,  r5,  r6,  lsr #16
        add             r8,  r7,  r8,  lsr #24
        mvn             r5,  r5
        mvn             r7,  r7
        tst             r6,  #0x100
        it              ne
        movne           r6,  r5,  lsr #24
        tst             r8,  #0x100
        it              ne
        movne           r8,  r7,  lsr #24
        orr             r9,  r9,  r6,  lsl #16
        ldr             r4,  [r1, #4]           /* moved form [B] */
        orr             r9,  r9,  r8,  lsl #24
        /* store dest */
        ldrsh           r5,  [r0, #8]           /* moved form [C] */
        str             r9,  [r1]

        /* load dest */
        /* [B] */
        /* block[4] and block[5] */
        /* [C] */
        ldrsh           r7,  [r0, #10]
        and             r6,  r4,  #0xFF
        and             r8,  r4,  #0xFF00
        add             r6,  r5,  r6
        add             r8,  r7,  r8,  lsr #8
        mvn             r5,  r5
        mvn             r7,  r7
        tst             r6,  #0x100
        it              ne
        movne           r6,  r5,  lsr #24
        tst             r8,  #0x100
        it              ne
        movne           r8,  r7,  lsr #24
        mov             r9,  r6
        ldrsh           r5,  [r0, #12]          /* moved from [D] */
        orr             r9,  r9,  r8,  lsl #8
        /* block[6] and block[7] */
        /* [D] */
        ldrsh           r7,  [r0, #14]
        and             r6,  r4,  #0xFF0000
        and             r8,  r4,  #0xFF000000
        add             r6,  r5,  r6,  lsr #16
        add             r8,  r7,  r8,  lsr #24
        mvn             r5,  r5
        mvn             r7,  r7
        tst             r6,  #0x100
        it              ne
        movne           r6,  r5,  lsr #24
        tst             r8,  #0x100
        it              ne
        movne           r8,  r7,  lsr #24
        orr             r9,  r9,  r6,  lsl #16
        add             r0,  r0,  #16           /* moved from [E] */
        orr             r9,  r9,  r8,  lsl #24
        subs            r10, r10, #1            /* moved from [F] */
        /* store dest */
        str             r9,  [r1, #4]

        /* [E] */
        /* [F] */
        add             r1,  r1,  r2
        bne             1b

        pop             {r4-r10}
        bx              lr
endfunc