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
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
|
//******************************************************************************
//*
//* Copyright (C) 2015 The Android Open Source Project
//*
//* Licensed under the Apache License, Version 2.0 (the "License");
//* you may not use this file except in compliance with the License.
//* You may obtain a copy of the License at:
//*
//* http://www.apache.org/licenses/LICENSE-2.0
//*
//* Unless required by applicable law or agreed to in writing, software
//* distributed under the License is distributed on an "AS IS" BASIS,
//* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
//* See the License for the specific language governing permissions and
//* limitations under the License.
//*
//*****************************************************************************
//* Originally developed and contributed by Ittiam Systems Pvt. Ltd, Bangalore
//*/
///**
//******************************************************************************
//* @file
//* ih264_inter_pred_luma_horz_hpel_vert_hpel_av8.s
//*
//* @brief
//* Contains function definitions for inter prediction interpolation.
//*
//* @author
//* Mohit
//*
//* @par List of Functions:
//*
//* - ih264_inter_pred_luma_horz_hpel_vert_hpel_av8()
//*
//* @remarks
//* None
//*
//*******************************************************************************
//*/
//void ih264_inter_pred_luma_horz_hpel_vert_hpel(UWORD8 *pu1_src,
// UWORD8 *pu1_dst,
// WORD32 src_strd,,
// WORD32 dst_strd,
// WORD32 ht,
// WORD32 wd,
// UWORD8* pu1_tmp,
// UWORD32 dydx)
//**************Variables Vs Registers*****************************************
// x0 => *pu1_src
// x1 => *pu1_dst
// x2 => src_strd
// x3 => dst_strd
// x4 => ht
// x5 => wd
.text
.p2align 2
.include "ih264_neon_macros.s"
.global ih264_inter_pred_luma_horz_hpel_vert_hpel_av8
ih264_inter_pred_luma_horz_hpel_vert_hpel_av8:
//store register values to stack
push_v_regs
stp x19, x20, [sp, #-16]!
sub x0, x0, x2, lsl #1 //pu1_src-2*src_strd
sub x0, x0, #2 //pu1_src-2
movi v26.8h, #0x14 // Filter coeff 20 into Q13
movi v24.8h, #0x5 // Filter coeff 5 into Q12
movi v27.8h, #0x14 // Filter coeff 20 into Q13
movi v25.8h, #0x5 // Filter coeff 5 into Q12
mov x7, #0x20
mov x8, #0x30
subs x12, x5, #4 //if wd=4 branch to loop_4
beq loop_4_start
subs x12, x5, #8 //if wd=8 branch to loop_8
beq loop_8_start
//when wd=16
movi v28.8h, #0x14 // Filter coeff 20 into Q13
movi v30.8h, #0x5 // Filter coeff 5 into Q12
sub x2, x2, #16
ld1 {v0.2s, v1.2s}, [x0], #16 // Vector load from src[0_0]
ld1 {v12.2s}, [x0], x2 // Vector load from src[0_0]
ld1 {v2.2s, v3.2s}, [x0], #16 // Vector load from src[1_0]
ld1 {v13.2s}, [x0], x2 // Vector load from src[1_0]
ld1 {v4.2s, v5.2s}, [x0], #16 // Vector load from src[2_0]
ld1 {v14.2s}, [x0], x2 // Vector load from src[2_0]
ld1 {v6.2s, v7.2s}, [x0], #16 // Vector load from src[3_0]
ld1 {v15.2s}, [x0], x2 // Vector load from src[3_0]
ld1 {v8.2s, v9.2s}, [x0], #16 // Vector load from src[4_0]
ld1 {v16.2s}, [x0], x2 // Vector load from src[4_0]
loop_16:
ld1 {v10.2s, v11.2s}, [x0], #16 // Vector load from src[5_0]
ld1 {v17.2s}, [x0], x2 // Vector load from src[5_0]
uaddl v20.8h, v4.8b, v6.8b
uaddl v18.8h, v0.8b, v10.8b
uaddl v22.8h, v2.8b, v8.8b
mla v18.8h, v20.8h , v28.8h
uaddl v24.8h, v5.8b, v7.8b
uaddl v20.8h, v1.8b, v11.8b
uaddl v26.8h, v3.8b, v9.8b
mla v20.8h, v24.8h , v28.8h
uaddl v24.8h, v14.8b, v15.8b
mls v18.8h, v22.8h , v30.8h
uaddl v22.8h, v12.8b, v17.8b
mls v20.8h, v26.8h , v30.8h
uaddl v26.8h, v13.8b, v16.8b
mla v22.8h, v24.8h , v28.8h
mls v22.8h, v26.8h , v30.8h
ext v24.16b, v18.16b , v20.16b , #4
ext v26.16b, v18.16b , v20.16b , #6
ext v23.16b, v18.16b , v20.16b , #10
add v0.8h, v24.8h , v26.8h
ext v24.16b, v18.16b , v20.16b , #2
ext v26.16b, v18.16b , v20.16b , #8
add v24.8h, v24.8h , v26.8h
saddl v26.4s, v18.4h, v23.4h
smlal v26.4s, v0.4h, v28.4h
smlsl v26.4s, v24.4h, v30.4h
saddl2 v23.4s, v18.8h, v23.8h
smlal2 v23.4s, v0.8h, v28.8h
smlsl2 v23.4s, v24.8h, v30.8h
sqrshrun v18.4h, v26.4s, #10
sqrshrun v19.4h, v23.4s, #10
uqxtn v18.8b, v18.8h
uqxtn v19.8b, v19.8h
mov v18.2s[1], v19.2s[0]
ext v24.16b, v20.16b , v22.16b , #4
ext v26.16b, v20.16b , v22.16b , #6
ext v0.16b, v20.16b , v22.16b , #10
add v25.8h, v24.8h , v26.8h
ext v24.16b, v20.16b , v22.16b , #2
ext v26.16b, v20.16b , v22.16b , #8
add v24.8h, v24.8h , v26.8h
saddl v26.4s, v0.4h, v20.4h
smlal v26.4s, v25.4h, v28.4h
smlsl v26.4s, v24.4h, v30.4h
saddl2 v22.4s, v0.8h, v20.8h
smlal2 v22.4s, v25.8h, v28.8h
smlsl2 v22.4s, v24.8h, v30.8h
sqrshrun v19.4h, v26.4s, #10
sqrshrun v25.4h, v22.4s, #10
uaddl v24.8h, v7.8b, v9.8b
uqxtn v19.8b, v19.8h
uqxtn v25.8b, v25.8h
mov v19.2s[1], v25.2s[0]
uaddl v22.8h, v4.8b, v10.8b
ld1 {v0.2s, v1.2s}, [x0], #16 // Vector load from src[6_0]
ld1 {v12.2s}, [x0], x2 // Vector load from src[6_0]
uaddl v20.8h, v6.8b, v8.8b
uaddl v26.8h, v5.8b, v11.8b
st1 {v18.2s, v19.2s}, [x1], x3 // store row 0
//ROW_2
uaddl v18.8h, v2.8b, v0.8b
mla v18.8h, v20.8h , v28.8h
uaddl v20.8h, v3.8b, v1.8b
mla v20.8h, v24.8h , v28.8h
uaddl v24.8h, v15.8b, v16.8b
mls v18.8h, v22.8h , v30.8h
uaddl v22.8h, v13.8b, v12.8b
mls v20.8h, v26.8h , v30.8h
uaddl v26.8h, v14.8b, v17.8b
mla v22.8h, v24.8h , v28.8h
mls v22.8h, v26.8h , v30.8h
ext v24.16b, v18.16b , v20.16b , #4
ext v26.16b, v18.16b , v20.16b , #6
ext v23.16b, v18.16b , v20.16b , #10
add v2.8h, v24.8h , v26.8h
ext v24.16b, v18.16b , v20.16b , #2
ext v26.16b, v18.16b , v20.16b , #8
add v24.8h, v24.8h , v26.8h
saddl v26.4s, v18.4h, v23.4h
smlal v26.4s, v2.4h, v28.4h
smlsl v26.4s, v24.4h, v30.4h
saddl2 v23.4s, v18.8h, v23.8h
smlal2 v23.4s, v2.8h, v28.8h
smlsl2 v23.4s, v24.8h, v30.8h
sqrshrun v18.4h, v26.4s, #10
sqrshrun v19.4h, v23.4s, #10
uqxtn v18.8b, v18.8h
uqxtn v19.8b, v19.8h
mov v18.2s[1], v19.2s[0]
ext v24.16b, v20.16b , v22.16b , #4
ext v26.16b, v20.16b , v22.16b , #6
ext v2.16b, v20.16b , v22.16b , #10
add v25.8h, v24.8h , v26.8h
ext v24.16b, v20.16b , v22.16b , #2
ext v26.16b, v20.16b , v22.16b , #8
add v24.8h, v24.8h , v26.8h
saddl v26.4s, v2.4h, v20.4h
smlal v26.4s, v25.4h, v28.4h
smlsl v26.4s, v24.4h, v30.4h
saddl2 v22.4s, v2.8h, v20.8h
smlal2 v22.4s, v25.8h, v28.8h
smlsl2 v22.4s, v24.8h, v30.8h
sqrshrun v19.4h, v26.4s, #10
sqrshrun v25.4h, v22.4s, #10
uaddl v24.8h, v9.8b, v11.8b
uqxtn v19.8b, v19.8h
uqxtn v25.8b, v25.8h
mov v19.2s[1], v25.2s[0]
uaddl v22.8h, v6.8b, v0.8b
ld1 {v2.2s, v3.2s}, [x0], #16 // Vector load from src[7_0]
ld1 {v13.2s}, [x0], x2 // Vector load from src[7_0]
uaddl v20.8h, v8.8b, v10.8b
uaddl v26.8h, v7.8b, v1.8b
st1 {v18.2s, v19.2s}, [x1], x3 // store row 1
//ROW_3
uaddl v18.8h, v4.8b, v2.8b
mla v18.8h, v20.8h , v28.8h
uaddl v20.8h, v5.8b, v3.8b
mla v20.8h, v24.8h , v28.8h
uaddl v24.8h, v16.8b, v17.8b
mls v18.8h, v22.8h , v30.8h
uaddl v22.8h, v14.8b, v13.8b
mls v20.8h, v26.8h , v30.8h
uaddl v26.8h, v15.8b, v12.8b
mla v22.8h, v24.8h , v28.8h
mls v22.8h, v26.8h , v30.8h
ext v24.16b, v18.16b , v20.16b , #4
ext v26.16b, v18.16b , v20.16b , #6
ext v23.16b, v18.16b , v20.16b , #10
add v4.8h, v24.8h , v26.8h
ext v24.16b, v18.16b , v20.16b , #2
ext v26.16b, v18.16b , v20.16b , #8
add v24.8h, v24.8h , v26.8h
saddl v26.4s, v18.4h, v23.4h
smlal v26.4s, v4.4h, v28.4h
smlsl v26.4s, v24.4h, v30.4h
saddl2 v23.4s, v18.8h, v23.8h
smlal2 v23.4s, v4.8h, v28.8h
smlsl2 v23.4s, v24.8h, v30.8h
sqrshrun v18.4h, v26.4s, #10
sqrshrun v19.4h, v23.4s, #10
uqxtn v18.8b, v18.8h
uqxtn v19.8b, v19.8h
mov v18.2s[1], v19.2s[0]
ext v24.16b, v20.16b , v22.16b , #4
ext v26.16b, v20.16b , v22.16b , #6
ext v4.16b, v20.16b , v22.16b , #10
add v25.8h, v24.8h , v26.8h
ext v24.16b, v20.16b , v22.16b , #2
ext v26.16b, v20.16b , v22.16b , #8
add v24.8h, v24.8h , v26.8h
saddl v26.4s, v4.4h, v20.4h
smlal v26.4s, v25.4h, v28.4h
smlsl v26.4s, v24.4h, v30.4h
saddl2 v22.4s, v4.8h, v20.8h
smlal2 v22.4s, v25.8h, v28.8h
smlsl2 v22.4s, v24.8h, v30.8h
sqrshrun v19.4h, v26.4s, #10
sqrshrun v25.4h, v22.4s, #10
uaddl v24.8h, v11.8b, v1.8b
uqxtn v19.8b, v19.8h
uqxtn v25.8b, v25.8h
mov v19.2s[1], v25.2s[0]
uaddl v22.8h, v8.8b, v2.8b
ld1 {v4.2s, v5.2s}, [x0], #16 // Vector load from src[8_0]
ld1 {v14.2s}, [x0], x2 // Vector load from src[8_0]
uaddl v20.8h, v10.8b, v0.8b
uaddl v26.8h, v9.8b, v3.8b
st1 {v18.2s, v19.2s}, [x1], x3 // store row 2
//ROW_4
uaddl v18.8h, v6.8b, v4.8b
mla v18.8h, v20.8h , v28.8h
uaddl v20.8h, v7.8b, v5.8b
mla v20.8h, v24.8h , v28.8h
uaddl v24.8h, v17.8b, v12.8b
mls v18.8h, v22.8h , v30.8h
uaddl v22.8h, v15.8b, v14.8b
mls v20.8h, v26.8h , v30.8h
uaddl v26.8h, v16.8b, v13.8b
mla v22.8h, v24.8h , v28.8h
mls v22.8h, v26.8h , v30.8h
ext v24.16b, v18.16b , v20.16b , #4
ext v26.16b, v18.16b , v20.16b , #6
ext v23.16b, v18.16b , v20.16b , #10
add v6.8h, v24.8h , v26.8h
ext v24.16b, v18.16b , v20.16b , #2
ext v26.16b, v18.16b , v20.16b , #8
add v24.8h, v24.8h , v26.8h
saddl v26.4s, v18.4h, v23.4h
smlal v26.4s, v6.4h, v28.4h
smlsl v26.4s, v24.4h, v30.4h
saddl2 v23.4s, v18.8h, v23.8h
smlal2 v23.4s, v6.8h, v28.8h
smlsl2 v23.4s, v24.8h, v30.8h
sqrshrun v18.4h, v26.4s, #10
sqrshrun v19.4h, v23.4s, #10
uqxtn v18.8b, v18.8h
uqxtn v19.8b, v19.8h
mov v18.2s[1], v19.2s[0]
ext v24.16b, v20.16b , v22.16b , #4
ext v26.16b, v20.16b , v22.16b , #6
ext v6.16b, v20.16b , v22.16b , #10
add v25.8h, v24.8h , v26.8h
ext v24.16b, v20.16b , v22.16b , #2
ext v26.16b, v20.16b , v22.16b , #8
add v24.8h, v24.8h , v26.8h
saddl v26.4s, v6.4h, v20.4h
smlal v26.4s, v25.4h, v28.4h
smlsl v26.4s, v24.4h, v30.4h
saddl2 v22.4s, v6.8h, v20.8h
smlal2 v22.4s, v25.8h, v28.8h
smlsl2 v22.4s, v24.8h, v30.8h
mov v6.16b, v2.16b
mov v7.16b, v3.16b
mov v2.16b, v10.16b
mov v3.16b, v11.16b
subs x4, x4, #4
sqrshrun v19.4h, v26.4s, #10
sqrshrun v25.4h, v22.4s, #10
mov v10.16b, v0.16b
mov v11.16b, v1.16b
mov v24.8b, v14.8b
mov v14.16b, v12.16b
mov v15.16b, v13.16b
uqxtn v19.8b, v19.8h
uqxtn v25.8b, v25.8h
mov v19.2s[1], v25.2s[0]
mov v0.16b, v8.16b
mov v1.16b, v9.16b
mov v8.16b, v4.16b
mov v9.16b, v5.16b
mov v12.16b, v16.16b
mov v13.16b, v17.16b
mov v4.16b, v10.16b
mov v5.16b, v11.16b
mov v16.8b, v24.8b
st1 {v18.2s, v19.2s}, [x1], x3 // store row 3
bgt loop_16 // looping if height =16
b end_func
loop_8_start:
ld1 {v0.2s, v1.2s}, [x0], x2 // Vector load from src[0_0]
ld1 {v2.2s, v3.2s}, [x0], x2 // Vector load from src[1_0]
ld1 {v4.2s, v5.2s}, [x0], x2 // Vector load from src[2_0]
ld1 {v6.2s, v7.2s}, [x0], x2 // Vector load from src[3_0]
ld1 {v8.2s, v9.2s}, [x0], x2 // Vector load from src[4_0]
loop_8:
ld1 {v10.2s, v11.2s}, [x0], x2 // Vector load from src[5_0]
uaddl v14.8h, v4.8b, v6.8b
uaddl v12.8h, v0.8b, v10.8b
uaddl v16.8h, v2.8b, v8.8b
mla v12.8h, v14.8h , v26.8h
uaddl v18.8h, v5.8b, v7.8b
uaddl v14.8h, v1.8b, v11.8b
uaddl v22.8h, v3.8b, v9.8b
mla v14.8h, v18.8h , v26.8h
mls v12.8h, v16.8h , v24.8h
ld1 {v0.2s, v1.2s}, [x0], x2 // Vector load from src[6_0]
uaddl v16.8h, v6.8b, v8.8b
mls v14.8h, v22.8h , v24.8h
uaddl v28.8h, v2.8b, v0.8b
ext v22.16b, v12.16b , v14.16b , #10
uaddl v18.8h, v4.8b, v10.8b
mla v28.8h, v16.8h , v26.8h
saddl v30.4s, v12.4h, v22.4h
saddl2 v22.4s, v12.8h, v22.8h
ext v16.16b, v12.16b , v14.16b , #4
mls v28.8h, v18.8h , v24.8h
ext v18.16b, v12.16b , v14.16b , #6
ext v20.16b, v12.16b , v14.16b , #8
ext v14.16b, v12.16b , v14.16b , #2
add v16.8h, v16.8h , v18.8h
add v18.8h, v14.8h , v20.8h
uaddl v20.8h, v7.8b, v9.8b
smlal v30.4s, v16.4h, v26.4h
smlsl v30.4s, v18.4h, v24.4h
smlal2 v22.4s, v16.8h, v26.8h
smlsl2 v22.4s, v18.8h, v24.8h
uaddl v14.8h, v3.8b, v1.8b
mla v14.8h, v20.8h , v26.8h
sqrshrun v12.4h, v30.4s, #10
uaddl v16.8h, v5.8b, v11.8b
sqrshrun v13.4h, v22.4s, #10
mls v14.8h, v16.8h , v24.8h
ld1 {v2.2s, v3.2s}, [x0], x2 // Vector load from src[7_0]
uqxtn v25.8b, v12.8h
uqxtn v13.8b, v13.8h
mov v25.2s[1], v13.2s[0]
uaddl v16.8h, v8.8b, v10.8b
ext v22.16b, v28.16b , v14.16b , #10
uaddl v20.8h, v4.8b, v2.8b
saddl v30.4s, v28.4h, v22.4h
mla v20.8h, v16.8h , v26.8h
saddl2 v22.4s, v28.8h, v22.8h
ext v16.16b, v28.16b , v14.16b , #4
ext v18.16b, v28.16b , v14.16b , #6
ext v12.16b, v28.16b , v14.16b , #8
ext v14.16b, v28.16b , v14.16b , #2
add v16.8h, v16.8h , v18.8h
add v18.8h, v12.8h , v14.8h
smlal v30.4s, v16.4h, v26.4h
smlsl v30.4s, v18.4h, v24.4h
smlal2 v22.4s, v16.8h, v26.8h
smlsl2 v22.4s, v18.8h, v24.8h
uaddl v18.8h, v6.8b, v0.8b
sqrshrun v16.4h, v30.4s, #10
sqrshrun v17.4h, v22.4s, #10
mov v12.8b, v25.8b
mov v25.8b, v24.8b
uaddl v28.8h, v9.8b, v11.8b
uqxtn v13.8b, v16.8h
uqxtn v17.8b, v17.8h
mov v13.2s[1], v17.2s[0]
uaddl v14.8h, v5.8b, v3.8b
uaddl v22.8h, v7.8b, v1.8b
mls v20.8h, v18.8h , v24.8h
st1 {v12.2s}, [x1], x3 // store row 0
mla v14.8h, v28.8h , v26.8h
ld1 {v4.2s, v5.2s}, [x0], x2 // Vector load from src[8_0]
uaddl v30.8h, v10.8b, v0.8b
uaddl v28.8h, v6.8b, v4.8b
mls v14.8h, v22.8h , v24.8h
st1 {v13.2s}, [x1], x3 // store row 1
mla v28.8h, v30.8h , v26.8h
ext v22.16b, v20.16b , v14.16b , #10
saddl v30.4s, v20.4h, v22.4h
saddl2 v22.4s, v20.8h, v22.8h
ext v16.16b, v20.16b , v14.16b , #4
ext v18.16b, v20.16b , v14.16b , #6
ext v12.16b, v20.16b , v14.16b , #8
ext v14.16b, v20.16b , v14.16b , #2
add v16.8h, v16.8h , v18.8h
add v18.8h, v14.8h , v12.8h
uaddl v20.8h, v8.8b, v2.8b
smlal v30.4s, v16.4h, v26.4h
smlsl v30.4s, v18.4h, v24.4h
smlal2 v22.4s, v16.8h, v26.8h
smlsl2 v22.4s, v18.8h, v24.8h
uaddl v18.8h, v11.8b, v1.8b
uaddl v16.8h, v7.8b, v5.8b
sqrshrun v12.4h, v30.4s, #10
uaddl v30.8h, v9.8b, v3.8b
mla v16.8h, v18.8h , v26.8h
sqrshrun v13.4h, v22.4s, #10
mls v28.8h, v20.8h , v24.8h
mls v16.8h, v30.8h , v24.8h
uqxtn v27.8b, v12.8h
uqxtn v13.8b, v13.8h
mov v27.2s[1], v13.2s[0]
ext v22.16b, v28.16b , v16.16b , #10
saddl v30.4s, v28.4h, v22.4h
saddl2 v22.4s, v28.8h, v22.8h
ext v12.16b, v28.16b , v16.16b , #4
ext v18.16b, v28.16b , v16.16b , #6
ext v20.16b, v28.16b , v16.16b , #8
ext v28.16b, v28.16b , v16.16b , #2
add v12.8h, v12.8h , v18.8h
add v18.8h, v28.8h , v20.8h
smlal v30.4s, v12.4h, v26.4h
smlsl v30.4s, v18.4h, v24.4h
smlal2 v22.4s, v12.8h, v26.8h
smlsl2 v22.4s, v18.8h, v24.8h
mov v12.8b, v27.8b
mov v27.8b, v26.8b
sqrshrun v16.4h, v30.4s, #10
mov v6.16b, v2.16b
mov v7.16b, v3.16b
sqrshrun v17.4h, v22.4s, #10
mov v2.16b, v10.16b
mov v3.16b, v11.16b
mov v10.16b, v0.16b
mov v11.16b, v1.16b
subs x4, x4, #4
uqxtn v13.8b, v16.8h
uqxtn v17.8b, v17.8h
mov v13.2s[1], v17.2s[0]
mov v0.16b, v8.16b
mov v1.16b, v9.16b
mov v8.16b, v4.16b
mov v9.16b, v5.16b
mov v4.16b, v10.16b
mov v5.16b, v11.16b
st1 {v12.2s}, [x1], x3 // store row 2
st1 {v13.2s}, [x1], x3 // store row 3
bgt loop_8 //if height =8 loop
b end_func
loop_4_start:
ld1 {v0.2s, v1.2s}, [x0], x2 // Vector load from src[0_0]
ld1 {v2.2s, v3.2s}, [x0], x2 // Vector load from src[1_0]
ld1 {v4.2s, v5.2s}, [x0], x2 // Vector load from src[2_0]
ld1 {v6.2s, v7.2s}, [x0], x2 // Vector load from src[3_0]
ld1 {v8.2s, v9.2s}, [x0], x2 // Vector load from src[4_0]
loop_4:
ld1 {v10.2s, v11.2s}, [x0], x2 // Vector load from src[5_0]
uaddl v14.8h, v4.8b, v6.8b // temp1 = src[2_0] + src[3_0]
uaddl v12.8h, v0.8b, v10.8b // temp = src[0_0] + src[5_0]
uaddl v16.8h, v2.8b, v8.8b // temp2 = src[1_0] + src[4_0]
mla v12.8h, v14.8h , v26.8h // temp += temp1 * 20
uaddl v18.8h, v5.8b, v7.8b // temp1 = src[2_0] + src[3_0]
uaddl v14.8h, v1.8b, v11.8b // temp = src[0_0] + src[5_0]
uaddl v22.8h, v3.8b, v9.8b // temp2 = src[1_0] + src[4_0]
mla v14.8h, v18.8h , v26.8h // temp += temp1 * 20
mls v12.8h, v16.8h , v24.8h // temp -= temp2 * 5
ld1 {v0.2s, v1.2s}, [x0], x2 // Vector load from src[6_0]
uaddl v16.8h, v6.8b, v8.8b
mls v14.8h, v22.8h , v24.8h // temp -= temp2 * 5
//Q6 and Q7 have filtered values
uaddl v28.8h, v2.8b, v0.8b
ext v22.16b, v12.16b , v14.16b , #10
uaddl v18.8h, v4.8b, v10.8b
mla v28.8h, v16.8h , v26.8h
saddl v30.4s, v12.4h, v22.4h
saddl v22.4s, v13.4h, v23.4h
ext v16.16b, v12.16b , v14.16b , #4
mls v28.8h, v18.8h , v24.8h
ext v18.16b, v12.16b , v14.16b , #6
ext v20.16b, v12.16b , v14.16b , #8
ext v14.16b, v12.16b , v14.16b , #2
add v16.8h, v16.8h , v18.8h
add v18.8h, v14.8h , v20.8h
uaddl v20.8h, v7.8b, v9.8b
smlal v30.4s, v16.4h, v26.4h
smlsl v30.4s, v18.4h, v24.4h
smlal v22.4s, v17.4h, v26.4h
smlsl v22.4s, v19.4h, v24.4h
uaddl v14.8h, v3.8b, v1.8b
mla v14.8h, v20.8h , v26.8h
sqrshrun v12.4h, v30.4s, #10
uaddl v16.8h, v5.8b, v11.8b
sqrshrun v13.4h, v22.4s, #10
mls v14.8h, v16.8h , v24.8h
ld1 {v2.2s, v3.2s}, [x0], x2 // Vector load from src[7_0]
uqxtn v25.8b, v12.8h
uaddl v16.8h, v8.8b, v10.8b
ext v22.16b, v28.16b , v14.16b , #10
uaddl v20.8h, v4.8b, v2.8b
saddl v30.4s, v28.4h, v22.4h
mla v20.8h, v16.8h , v26.8h
saddl v22.4s, v29.4h, v23.4h
ext v16.16b, v28.16b , v14.16b , #4
ext v18.16b, v28.16b , v14.16b , #6
ext v12.16b, v28.16b , v14.16b , #8
ext v14.16b, v28.16b , v14.16b , #2
add v16.8h, v16.8h , v18.8h
add v18.8h, v12.8h , v14.8h
smlal v30.4s, v16.4h, v26.4h
smlsl v30.4s, v18.4h, v24.4h
smlal v22.4s, v17.4h, v26.4h
smlsl v22.4s, v19.4h, v24.4h
uaddl v18.8h, v6.8b, v0.8b
sqrshrun v16.4h, v30.4s, #10
sqrshrun v17.4h, v22.4s, #10
mov v12.8b, v25.8b
mov v25.8b, v24.8b
uaddl v28.8h, v9.8b, v11.8b
uqxtn v13.8b, v16.8h
uaddl v14.8h, v5.8b, v3.8b
uaddl v22.8h, v7.8b, v1.8b
mls v20.8h, v18.8h , v24.8h
st1 {v12.s}[0], [x1], x3 // store row 0
mla v14.8h, v28.8h , v26.8h
ld1 {v4.2s, v5.2s}, [x0], x2 // Vector load from src[8_0]
uaddl v30.8h, v10.8b, v0.8b
uaddl v28.8h, v6.8b, v4.8b
mls v14.8h, v22.8h , v24.8h
st1 {v13.s}[0], [x1], x3 //store row 1
mla v28.8h, v30.8h , v26.8h
ext v22.16b, v20.16b , v14.16b , #10
saddl v30.4s, v20.4h, v22.4h
saddl v22.4s, v21.4h, v23.4h
ext v16.16b, v20.16b , v14.16b , #4
ext v18.16b, v20.16b , v14.16b , #6
ext v12.16b, v20.16b , v14.16b , #8
ext v14.16b, v20.16b , v14.16b , #2
add v16.8h, v16.8h , v18.8h
add v18.8h, v14.8h , v12.8h
uaddl v20.8h, v8.8b, v2.8b
smlal v30.4s, v16.4h, v26.4h
smlsl v30.4s, v18.4h, v24.4h
smlal v22.4s, v17.4h, v26.4h
smlsl v22.4s, v19.4h, v24.4h
uaddl v18.8h, v11.8b, v1.8b
uaddl v16.8h, v7.8b, v5.8b
sqrshrun v12.4h, v30.4s, #10
uaddl v30.8h, v9.8b, v3.8b
mla v16.8h, v18.8h , v26.8h
sqrshrun v13.4h, v22.4s, #10
mls v28.8h, v20.8h , v24.8h
mls v16.8h, v30.8h , v24.8h
uqxtn v27.8b, v12.8h
ext v22.16b, v28.16b , v16.16b , #10
saddl v30.4s, v28.4h, v22.4h
saddl v22.4s, v29.4h, v23.4h
ext v12.16b, v28.16b , v16.16b , #4
ext v18.16b, v28.16b , v16.16b , #6
ext v20.16b, v28.16b , v16.16b , #8
ext v28.16b, v28.16b , v16.16b , #2
add v12.8h, v12.8h , v18.8h
add v18.8h, v28.8h , v20.8h
smlal v30.4s, v12.4h, v26.4h
smlsl v30.4s, v18.4h, v24.4h
smlal v22.4s, v13.4h, v26.4h
smlsl v22.4s, v19.4h, v24.4h
mov v12.8b, v27.8b
mov v27.8b, v26.8b
sqrshrun v16.4h, v30.4s, #10
mov v6.16b, v2.16b
mov v7.16b, v3.16b
sqrshrun v17.4h, v22.4s, #10
mov v2.16b, v10.16b
mov v3.16b, v11.16b
mov v10.16b, v0.16b
mov v11.16b, v1.16b
subs x4, x4, #4
uqxtn v13.8b, v16.8h
mov v0.16b, v8.16b
mov v1.16b, v9.16b
mov v8.16b, v4.16b
mov v9.16b, v5.16b
mov v4.16b, v10.16b
mov v5.16b, v11.16b
st1 {v12.s}[0], [x1], x3 // store row 2
st1 {v13.s}[0], [x1], x3 // store row 3
bgt loop_4
end_func:
//Restoring registers from stack
ldp x19, x20, [sp], #16
pop_v_regs
ret
|