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
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
|
/*
* Copyright 2012 The LibYuv Project Authors. All rights reserved.
*
* Use of this source code is governed by a BSD-style license
* that can be found in the LICENSE file in the root of the source
* tree. An additional intellectual property rights grant can be found
* in the file PATENTS. All contributing project authors may
* be found in the AUTHORS file in the root of the source tree.
*/
#include "libyuv/row.h"
#include <string.h> // For memset.
#include "libyuv/basic_types.h"
#ifdef __cplusplus
namespace libyuv {
extern "C" {
#endif
// memset for temp is meant to clear the source buffer (not dest) so that
// SIMD that reads full multiple of 16 bytes will not trigger msan errors.
// memset is not needed for production, as the garbage values are processed but
// not used, although there may be edge cases for subsampling.
// The size of the buffer is based on the largest read, which can be inferred
// by the source type (e.g. ARGB) and the mask (last parameter), or by examining
// the source code for how much the source pointers are advanced.
// Subsampled source needs to be increase by 1 of not even.
#define SS(width, shift) (((width) + (1 << (shift)) - 1) >> (shift))
// Any 4 planes to 1 with yuvconstants
#define ANY41C(NAMEANY, ANY_SIMD, UVSHIFT, DUVSHIFT, BPP, MASK) \
void NAMEANY(const uint8_t* y_buf, const uint8_t* u_buf, \
const uint8_t* v_buf, const uint8_t* a_buf, uint8_t* dst_ptr, \
const struct YuvConstants* yuvconstants, int width) { \
SIMD_ALIGNED(uint8_t temp[64 * 5]); \
memset(temp, 0, 64 * 4); /* for msan */ \
int r = width & MASK; \
int n = width & ~MASK; \
if (n > 0) { \
ANY_SIMD(y_buf, u_buf, v_buf, a_buf, dst_ptr, yuvconstants, n); \
} \
memcpy(temp, y_buf + n, r); \
memcpy(temp + 64, u_buf + (n >> UVSHIFT), SS(r, UVSHIFT)); \
memcpy(temp + 128, v_buf + (n >> UVSHIFT), SS(r, UVSHIFT)); \
memcpy(temp + 192, a_buf + n, r); \
ANY_SIMD(temp, temp + 64, temp + 128, temp + 192, temp + 256, \
yuvconstants, MASK + 1); \
memcpy(dst_ptr + (n >> DUVSHIFT) * BPP, temp + 256, \
SS(r, DUVSHIFT) * BPP); \
}
#ifdef HAS_I422ALPHATOARGBROW_SSSE3
ANY41C(I422AlphaToARGBRow_Any_SSSE3, I422AlphaToARGBRow_SSSE3, 1, 0, 4, 7)
#endif
#ifdef HAS_I422ALPHATOARGBROW_AVX2
ANY41C(I422AlphaToARGBRow_Any_AVX2, I422AlphaToARGBRow_AVX2, 1, 0, 4, 15)
#endif
#ifdef HAS_I422ALPHATOARGBROW_NEON
ANY41C(I422AlphaToARGBRow_Any_NEON, I422AlphaToARGBRow_NEON, 1, 0, 4, 7)
#endif
#ifdef HAS_I422ALPHATOARGBROW_MSA
ANY41C(I422AlphaToARGBRow_Any_MSA, I422AlphaToARGBRow_MSA, 1, 0, 4, 7)
#endif
#ifdef HAS_I422ALPHATOARGBROW_MMI
ANY41C(I422AlphaToARGBRow_Any_MMI, I422AlphaToARGBRow_MMI, 1, 0, 4, 7)
#endif
#undef ANY41C
// Any 3 planes to 1.
#define ANY31(NAMEANY, ANY_SIMD, UVSHIFT, DUVSHIFT, BPP, MASK) \
void NAMEANY(const uint8_t* y_buf, const uint8_t* u_buf, \
const uint8_t* v_buf, uint8_t* dst_ptr, int width) { \
SIMD_ALIGNED(uint8_t temp[64 * 4]); \
memset(temp, 0, 64 * 3); /* for YUY2 and msan */ \
int r = width & MASK; \
int n = width & ~MASK; \
if (n > 0) { \
ANY_SIMD(y_buf, u_buf, v_buf, dst_ptr, n); \
} \
memcpy(temp, y_buf + n, r); \
memcpy(temp + 64, u_buf + (n >> UVSHIFT), SS(r, UVSHIFT)); \
memcpy(temp + 128, v_buf + (n >> UVSHIFT), SS(r, UVSHIFT)); \
ANY_SIMD(temp, temp + 64, temp + 128, temp + 192, MASK + 1); \
memcpy(dst_ptr + (n >> DUVSHIFT) * BPP, temp + 192, \
SS(r, DUVSHIFT) * BPP); \
}
// Merge functions.
#ifdef HAS_MERGERGBROW_SSSE3
ANY31(MergeRGBRow_Any_SSSE3, MergeRGBRow_SSSE3, 0, 0, 3, 15)
#endif
#ifdef HAS_MERGERGBROW_NEON
ANY31(MergeRGBRow_Any_NEON, MergeRGBRow_NEON, 0, 0, 3, 15)
#endif
#ifdef HAS_MERGERGBROW_MMI
ANY31(MergeRGBRow_Any_MMI, MergeRGBRow_MMI, 0, 0, 3, 7)
#endif
#ifdef HAS_I422TOYUY2ROW_SSE2
ANY31(I422ToYUY2Row_Any_SSE2, I422ToYUY2Row_SSE2, 1, 1, 4, 15)
ANY31(I422ToUYVYRow_Any_SSE2, I422ToUYVYRow_SSE2, 1, 1, 4, 15)
#endif
#ifdef HAS_I422TOYUY2ROW_AVX2
ANY31(I422ToYUY2Row_Any_AVX2, I422ToYUY2Row_AVX2, 1, 1, 4, 31)
ANY31(I422ToUYVYRow_Any_AVX2, I422ToUYVYRow_AVX2, 1, 1, 4, 31)
#endif
#ifdef HAS_I422TOYUY2ROW_NEON
ANY31(I422ToYUY2Row_Any_NEON, I422ToYUY2Row_NEON, 1, 1, 4, 15)
#endif
#ifdef HAS_I422TOYUY2ROW_MSA
ANY31(I422ToYUY2Row_Any_MSA, I422ToYUY2Row_MSA, 1, 1, 4, 31)
#endif
#ifdef HAS_I422TOYUY2ROW_MMI
ANY31(I422ToYUY2Row_Any_MMI, I422ToYUY2Row_MMI, 1, 1, 4, 7)
#endif
#ifdef HAS_I422TOUYVYROW_NEON
ANY31(I422ToUYVYRow_Any_NEON, I422ToUYVYRow_NEON, 1, 1, 4, 15)
#endif
#ifdef HAS_I422TOUYVYROW_MSA
ANY31(I422ToUYVYRow_Any_MSA, I422ToUYVYRow_MSA, 1, 1, 4, 31)
#endif
#ifdef HAS_I422TOUYVYROW_MMI
ANY31(I422ToUYVYRow_Any_MMI, I422ToUYVYRow_MMI, 1, 1, 4, 7)
#endif
#ifdef HAS_BLENDPLANEROW_AVX2
ANY31(BlendPlaneRow_Any_AVX2, BlendPlaneRow_AVX2, 0, 0, 1, 31)
#endif
#ifdef HAS_BLENDPLANEROW_SSSE3
ANY31(BlendPlaneRow_Any_SSSE3, BlendPlaneRow_SSSE3, 0, 0, 1, 7)
#endif
#ifdef HAS_BLENDPLANEROW_MMI
ANY31(BlendPlaneRow_Any_MMI, BlendPlaneRow_MMI, 0, 0, 1, 7)
#endif
#undef ANY31
// Note that odd width replication includes 444 due to implementation
// on arm that subsamples 444 to 422 internally.
// Any 3 planes to 1 with yuvconstants
#define ANY31C(NAMEANY, ANY_SIMD, UVSHIFT, DUVSHIFT, BPP, MASK) \
void NAMEANY(const uint8_t* y_buf, const uint8_t* u_buf, \
const uint8_t* v_buf, uint8_t* dst_ptr, \
const struct YuvConstants* yuvconstants, int width) { \
SIMD_ALIGNED(uint8_t temp[128 * 4]); \
memset(temp, 0, 128 * 3); /* for YUY2 and msan */ \
int r = width & MASK; \
int n = width & ~MASK; \
if (n > 0) { \
ANY_SIMD(y_buf, u_buf, v_buf, dst_ptr, yuvconstants, n); \
} \
memcpy(temp, y_buf + n, r); \
memcpy(temp + 128, u_buf + (n >> UVSHIFT), SS(r, UVSHIFT)); \
memcpy(temp + 256, v_buf + (n >> UVSHIFT), SS(r, UVSHIFT)); \
if (width & 1) { \
temp[128 + SS(r, UVSHIFT)] = temp[128 + SS(r, UVSHIFT) - 1]; \
temp[256 + SS(r, UVSHIFT)] = temp[256 + SS(r, UVSHIFT) - 1]; \
} \
ANY_SIMD(temp, temp + 128, temp + 256, temp + 384, yuvconstants, \
MASK + 1); \
memcpy(dst_ptr + (n >> DUVSHIFT) * BPP, temp + 384, \
SS(r, DUVSHIFT) * BPP); \
}
#ifdef HAS_I422TOARGBROW_SSSE3
ANY31C(I422ToARGBRow_Any_SSSE3, I422ToARGBRow_SSSE3, 1, 0, 4, 7)
#endif
#ifdef HAS_I422TOAR30ROW_SSSE3
ANY31C(I422ToAR30Row_Any_SSSE3, I422ToAR30Row_SSSE3, 1, 0, 4, 7)
#endif
#ifdef HAS_I422TOAR30ROW_AVX2
ANY31C(I422ToAR30Row_Any_AVX2, I422ToAR30Row_AVX2, 1, 0, 4, 15)
#endif
#ifdef HAS_I444TOARGBROW_SSSE3
ANY31C(I444ToARGBRow_Any_SSSE3, I444ToARGBRow_SSSE3, 0, 0, 4, 7)
ANY31C(I422ToRGBARow_Any_SSSE3, I422ToRGBARow_SSSE3, 1, 0, 4, 7)
ANY31C(I422ToARGB4444Row_Any_SSSE3, I422ToARGB4444Row_SSSE3, 1, 0, 2, 7)
ANY31C(I422ToARGB1555Row_Any_SSSE3, I422ToARGB1555Row_SSSE3, 1, 0, 2, 7)
ANY31C(I422ToRGB565Row_Any_SSSE3, I422ToRGB565Row_SSSE3, 1, 0, 2, 7)
ANY31C(I422ToRGB24Row_Any_SSSE3, I422ToRGB24Row_SSSE3, 1, 0, 3, 15)
#endif // HAS_I444TOARGBROW_SSSE3
#ifdef HAS_I422TORGB24ROW_AVX2
ANY31C(I422ToRGB24Row_Any_AVX2, I422ToRGB24Row_AVX2, 1, 0, 3, 31)
#endif
#ifdef HAS_I422TOARGBROW_AVX2
ANY31C(I422ToARGBRow_Any_AVX2, I422ToARGBRow_AVX2, 1, 0, 4, 15)
#endif
#ifdef HAS_I422TORGBAROW_AVX2
ANY31C(I422ToRGBARow_Any_AVX2, I422ToRGBARow_AVX2, 1, 0, 4, 15)
#endif
#ifdef HAS_I444TOARGBROW_AVX2
ANY31C(I444ToARGBRow_Any_AVX2, I444ToARGBRow_AVX2, 0, 0, 4, 15)
#endif
#ifdef HAS_I422TOARGB4444ROW_AVX2
ANY31C(I422ToARGB4444Row_Any_AVX2, I422ToARGB4444Row_AVX2, 1, 0, 2, 15)
#endif
#ifdef HAS_I422TOARGB1555ROW_AVX2
ANY31C(I422ToARGB1555Row_Any_AVX2, I422ToARGB1555Row_AVX2, 1, 0, 2, 15)
#endif
#ifdef HAS_I422TORGB565ROW_AVX2
ANY31C(I422ToRGB565Row_Any_AVX2, I422ToRGB565Row_AVX2, 1, 0, 2, 15)
#endif
#ifdef HAS_I422TOARGBROW_NEON
ANY31C(I444ToARGBRow_Any_NEON, I444ToARGBRow_NEON, 0, 0, 4, 7)
ANY31C(I422ToARGBRow_Any_NEON, I422ToARGBRow_NEON, 1, 0, 4, 7)
ANY31C(I422ToRGBARow_Any_NEON, I422ToRGBARow_NEON, 1, 0, 4, 7)
ANY31C(I422ToRGB24Row_Any_NEON, I422ToRGB24Row_NEON, 1, 0, 3, 7)
ANY31C(I422ToARGB4444Row_Any_NEON, I422ToARGB4444Row_NEON, 1, 0, 2, 7)
ANY31C(I422ToARGB1555Row_Any_NEON, I422ToARGB1555Row_NEON, 1, 0, 2, 7)
ANY31C(I422ToRGB565Row_Any_NEON, I422ToRGB565Row_NEON, 1, 0, 2, 7)
#endif
#ifdef HAS_I422TOARGBROW_MSA
ANY31C(I444ToARGBRow_Any_MSA, I444ToARGBRow_MSA, 0, 0, 4, 7)
ANY31C(I422ToARGBRow_Any_MSA, I422ToARGBRow_MSA, 1, 0, 4, 7)
ANY31C(I422ToRGBARow_Any_MSA, I422ToRGBARow_MSA, 1, 0, 4, 7)
ANY31C(I422ToRGB24Row_Any_MSA, I422ToRGB24Row_MSA, 1, 0, 3, 15)
ANY31C(I422ToARGB4444Row_Any_MSA, I422ToARGB4444Row_MSA, 1, 0, 2, 7)
ANY31C(I422ToARGB1555Row_Any_MSA, I422ToARGB1555Row_MSA, 1, 0, 2, 7)
ANY31C(I422ToRGB565Row_Any_MSA, I422ToRGB565Row_MSA, 1, 0, 2, 7)
#endif
#ifdef HAS_I422TOARGBROW_MMI
ANY31C(I444ToARGBRow_Any_MMI, I444ToARGBRow_MMI, 0, 0, 4, 7)
ANY31C(I422ToARGBRow_Any_MMI, I422ToARGBRow_MMI, 1, 0, 4, 7)
ANY31C(I422ToRGB24Row_Any_MMI, I422ToRGB24Row_MMI, 1, 0, 3, 15)
ANY31C(I422ToARGB4444Row_Any_MMI, I422ToARGB4444Row_MMI, 1, 0, 2, 7)
ANY31C(I422ToARGB1555Row_Any_MMI, I422ToARGB1555Row_MMI, 1, 0, 2, 7)
ANY31C(I422ToRGB565Row_Any_MMI, I422ToRGB565Row_MMI, 1, 0, 2, 7)
ANY31C(I422ToRGBARow_Any_MMI, I422ToRGBARow_MMI, 1, 0, 4, 7)
#endif
#undef ANY31C
// Any 3 planes of 16 bit to 1 with yuvconstants
// TODO(fbarchard): consider sharing this code with ANY31C
#define ANY31CT(NAMEANY, ANY_SIMD, UVSHIFT, DUVSHIFT, T, SBPP, BPP, MASK) \
void NAMEANY(const T* y_buf, const T* u_buf, const T* v_buf, \
uint8_t* dst_ptr, const struct YuvConstants* yuvconstants, \
int width) { \
SIMD_ALIGNED(T temp[16 * 3]); \
SIMD_ALIGNED(uint8_t out[64]); \
memset(temp, 0, 16 * 3 * SBPP); /* for YUY2 and msan */ \
int r = width & MASK; \
int n = width & ~MASK; \
if (n > 0) { \
ANY_SIMD(y_buf, u_buf, v_buf, dst_ptr, yuvconstants, n); \
} \
memcpy(temp, y_buf + n, r * SBPP); \
memcpy(temp + 16, u_buf + (n >> UVSHIFT), SS(r, UVSHIFT) * SBPP); \
memcpy(temp + 32, v_buf + (n >> UVSHIFT), SS(r, UVSHIFT) * SBPP); \
ANY_SIMD(temp, temp + 16, temp + 32, out, yuvconstants, MASK + 1); \
memcpy(dst_ptr + (n >> DUVSHIFT) * BPP, out, SS(r, DUVSHIFT) * BPP); \
}
#ifdef HAS_I210TOAR30ROW_SSSE3
ANY31CT(I210ToAR30Row_Any_SSSE3, I210ToAR30Row_SSSE3, 1, 0, uint16_t, 2, 4, 7)
#endif
#ifdef HAS_I210TOARGBROW_SSSE3
ANY31CT(I210ToARGBRow_Any_SSSE3, I210ToARGBRow_SSSE3, 1, 0, uint16_t, 2, 4, 7)
#endif
#ifdef HAS_I210TOARGBROW_AVX2
ANY31CT(I210ToARGBRow_Any_AVX2, I210ToARGBRow_AVX2, 1, 0, uint16_t, 2, 4, 15)
#endif
#ifdef HAS_I210TOAR30ROW_AVX2
ANY31CT(I210ToAR30Row_Any_AVX2, I210ToAR30Row_AVX2, 1, 0, uint16_t, 2, 4, 15)
#endif
#ifdef HAS_I210TOARGBROW_MMI
ANY31CT(I210ToARGBRow_Any_MMI, I210ToARGBRow_MMI, 1, 0, uint16_t, 2, 4, 7)
#endif
#undef ANY31CT
// Any 2 planes to 1.
#define ANY21(NAMEANY, ANY_SIMD, UVSHIFT, SBPP, SBPP2, BPP, MASK) \
void NAMEANY(const uint8_t* y_buf, const uint8_t* uv_buf, uint8_t* dst_ptr, \
int width) { \
SIMD_ALIGNED(uint8_t temp[64 * 3]); \
memset(temp, 0, 64 * 2); /* for msan */ \
int r = width & MASK; \
int n = width & ~MASK; \
if (n > 0) { \
ANY_SIMD(y_buf, uv_buf, dst_ptr, n); \
} \
memcpy(temp, y_buf + n * SBPP, r * SBPP); \
memcpy(temp + 64, uv_buf + (n >> UVSHIFT) * SBPP2, \
SS(r, UVSHIFT) * SBPP2); \
ANY_SIMD(temp, temp + 64, temp + 128, MASK + 1); \
memcpy(dst_ptr + n * BPP, temp + 128, r * BPP); \
}
// Merge functions.
#ifdef HAS_MERGEUVROW_SSE2
ANY21(MergeUVRow_Any_SSE2, MergeUVRow_SSE2, 0, 1, 1, 2, 15)
#endif
#ifdef HAS_MERGEUVROW_AVX2
ANY21(MergeUVRow_Any_AVX2, MergeUVRow_AVX2, 0, 1, 1, 2, 31)
#endif
#ifdef HAS_MERGEUVROW_NEON
ANY21(MergeUVRow_Any_NEON, MergeUVRow_NEON, 0, 1, 1, 2, 15)
#endif
#ifdef HAS_MERGEUVROW_MSA
ANY21(MergeUVRow_Any_MSA, MergeUVRow_MSA, 0, 1, 1, 2, 15)
#endif
#ifdef HAS_MERGEUVROW_MMI
ANY21(MergeUVRow_Any_MMI, MergeUVRow_MMI, 0, 1, 1, 2, 7)
#endif
#ifdef HAS_NV21TOYUV24ROW_NEON
ANY21(NV21ToYUV24Row_Any_NEON, NV21ToYUV24Row_NEON, 1, 1, 2, 3, 15)
#endif
#ifdef HAS_NV21TOYUV24ROW_AVX2
ANY21(NV21ToYUV24Row_Any_AVX2, NV21ToYUV24Row_AVX2, 1, 1, 2, 3, 31)
#endif
// Math functions.
#ifdef HAS_ARGBMULTIPLYROW_SSE2
ANY21(ARGBMultiplyRow_Any_SSE2, ARGBMultiplyRow_SSE2, 0, 4, 4, 4, 3)
#endif
#ifdef HAS_ARGBADDROW_SSE2
ANY21(ARGBAddRow_Any_SSE2, ARGBAddRow_SSE2, 0, 4, 4, 4, 3)
#endif
#ifdef HAS_ARGBSUBTRACTROW_SSE2
ANY21(ARGBSubtractRow_Any_SSE2, ARGBSubtractRow_SSE2, 0, 4, 4, 4, 3)
#endif
#ifdef HAS_ARGBMULTIPLYROW_AVX2
ANY21(ARGBMultiplyRow_Any_AVX2, ARGBMultiplyRow_AVX2, 0, 4, 4, 4, 7)
#endif
#ifdef HAS_ARGBADDROW_AVX2
ANY21(ARGBAddRow_Any_AVX2, ARGBAddRow_AVX2, 0, 4, 4, 4, 7)
#endif
#ifdef HAS_ARGBSUBTRACTROW_AVX2
ANY21(ARGBSubtractRow_Any_AVX2, ARGBSubtractRow_AVX2, 0, 4, 4, 4, 7)
#endif
#ifdef HAS_ARGBMULTIPLYROW_NEON
ANY21(ARGBMultiplyRow_Any_NEON, ARGBMultiplyRow_NEON, 0, 4, 4, 4, 7)
#endif
#ifdef HAS_ARGBADDROW_NEON
ANY21(ARGBAddRow_Any_NEON, ARGBAddRow_NEON, 0, 4, 4, 4, 7)
#endif
#ifdef HAS_ARGBSUBTRACTROW_NEON
ANY21(ARGBSubtractRow_Any_NEON, ARGBSubtractRow_NEON, 0, 4, 4, 4, 7)
#endif
#ifdef HAS_ARGBMULTIPLYROW_MSA
ANY21(ARGBMultiplyRow_Any_MSA, ARGBMultiplyRow_MSA, 0, 4, 4, 4, 3)
#endif
#ifdef HAS_ARGBMULTIPLYROW_MMI
ANY21(ARGBMultiplyRow_Any_MMI, ARGBMultiplyRow_MMI, 0, 4, 4, 4, 1)
#endif
#ifdef HAS_ARGBADDROW_MSA
ANY21(ARGBAddRow_Any_MSA, ARGBAddRow_MSA, 0, 4, 4, 4, 7)
#endif
#ifdef HAS_ARGBADDROW_MMI
ANY21(ARGBAddRow_Any_MMI, ARGBAddRow_MMI, 0, 4, 4, 4, 1)
#endif
#ifdef HAS_ARGBSUBTRACTROW_MSA
ANY21(ARGBSubtractRow_Any_MSA, ARGBSubtractRow_MSA, 0, 4, 4, 4, 7)
#endif
#ifdef HAS_ARGBSUBTRACTROW_MMI
ANY21(ARGBSubtractRow_Any_MMI, ARGBSubtractRow_MMI, 0, 4, 4, 4, 1)
#endif
#ifdef HAS_SOBELROW_SSE2
ANY21(SobelRow_Any_SSE2, SobelRow_SSE2, 0, 1, 1, 4, 15)
#endif
#ifdef HAS_SOBELROW_NEON
ANY21(SobelRow_Any_NEON, SobelRow_NEON, 0, 1, 1, 4, 7)
#endif
#ifdef HAS_SOBELROW_MSA
ANY21(SobelRow_Any_MSA, SobelRow_MSA, 0, 1, 1, 4, 15)
#endif
#ifdef HAS_SOBELROW_MMI
ANY21(SobelRow_Any_MMI, SobelRow_MMI, 0, 1, 1, 4, 7)
#endif
#ifdef HAS_SOBELTOPLANEROW_SSE2
ANY21(SobelToPlaneRow_Any_SSE2, SobelToPlaneRow_SSE2, 0, 1, 1, 1, 15)
#endif
#ifdef HAS_SOBELTOPLANEROW_NEON
ANY21(SobelToPlaneRow_Any_NEON, SobelToPlaneRow_NEON, 0, 1, 1, 1, 15)
#endif
#ifdef HAS_SOBELTOPLANEROW_MSA
ANY21(SobelToPlaneRow_Any_MSA, SobelToPlaneRow_MSA, 0, 1, 1, 1, 31)
#endif
#ifdef HAS_SOBELTOPLANEROW_MMI
ANY21(SobelToPlaneRow_Any_MMI, SobelToPlaneRow_MMI, 0, 1, 1, 1, 7)
#endif
#ifdef HAS_SOBELXYROW_SSE2
ANY21(SobelXYRow_Any_SSE2, SobelXYRow_SSE2, 0, 1, 1, 4, 15)
#endif
#ifdef HAS_SOBELXYROW_NEON
ANY21(SobelXYRow_Any_NEON, SobelXYRow_NEON, 0, 1, 1, 4, 7)
#endif
#ifdef HAS_SOBELXYROW_MSA
ANY21(SobelXYRow_Any_MSA, SobelXYRow_MSA, 0, 1, 1, 4, 15)
#endif
#ifdef HAS_SOBELXYROW_MMI
ANY21(SobelXYRow_Any_MMI, SobelXYRow_MMI, 0, 1, 1, 4, 7)
#endif
#undef ANY21
// Any 2 planes to 1 with yuvconstants
#define ANY21C(NAMEANY, ANY_SIMD, UVSHIFT, SBPP, SBPP2, BPP, MASK) \
void NAMEANY(const uint8_t* y_buf, const uint8_t* uv_buf, uint8_t* dst_ptr, \
const struct YuvConstants* yuvconstants, int width) { \
SIMD_ALIGNED(uint8_t temp[128 * 3]); \
memset(temp, 0, 128 * 2); /* for msan */ \
int r = width & MASK; \
int n = width & ~MASK; \
if (n > 0) { \
ANY_SIMD(y_buf, uv_buf, dst_ptr, yuvconstants, n); \
} \
memcpy(temp, y_buf + n * SBPP, r * SBPP); \
memcpy(temp + 128, uv_buf + (n >> UVSHIFT) * SBPP2, \
SS(r, UVSHIFT) * SBPP2); \
ANY_SIMD(temp, temp + 128, temp + 256, yuvconstants, MASK + 1); \
memcpy(dst_ptr + n * BPP, temp + 256, r * BPP); \
}
// Biplanar to RGB.
#ifdef HAS_NV12TOARGBROW_SSSE3
ANY21C(NV12ToARGBRow_Any_SSSE3, NV12ToARGBRow_SSSE3, 1, 1, 2, 4, 7)
#endif
#ifdef HAS_NV12TOARGBROW_AVX2
ANY21C(NV12ToARGBRow_Any_AVX2, NV12ToARGBRow_AVX2, 1, 1, 2, 4, 15)
#endif
#ifdef HAS_NV12TOARGBROW_NEON
ANY21C(NV12ToARGBRow_Any_NEON, NV12ToARGBRow_NEON, 1, 1, 2, 4, 7)
#endif
#ifdef HAS_NV12TOARGBROW_MSA
ANY21C(NV12ToARGBRow_Any_MSA, NV12ToARGBRow_MSA, 1, 1, 2, 4, 7)
#endif
#ifdef HAS_NV12TOARGBROW_MMI
ANY21C(NV12ToARGBRow_Any_MMI, NV12ToARGBRow_MMI, 1, 1, 2, 4, 7)
#endif
#ifdef HAS_NV21TOARGBROW_SSSE3
ANY21C(NV21ToARGBRow_Any_SSSE3, NV21ToARGBRow_SSSE3, 1, 1, 2, 4, 7)
#endif
#ifdef HAS_NV21TOARGBROW_AVX2
ANY21C(NV21ToARGBRow_Any_AVX2, NV21ToARGBRow_AVX2, 1, 1, 2, 4, 15)
#endif
#ifdef HAS_NV21TOARGBROW_NEON
ANY21C(NV21ToARGBRow_Any_NEON, NV21ToARGBRow_NEON, 1, 1, 2, 4, 7)
#endif
#ifdef HAS_NV21TOARGBROW_MSA
ANY21C(NV21ToARGBRow_Any_MSA, NV21ToARGBRow_MSA, 1, 1, 2, 4, 7)
#endif
#ifdef HAS_NV21TOARGBROW_MMI
ANY21C(NV21ToARGBRow_Any_MMI, NV21ToARGBRow_MMI, 1, 1, 2, 4, 7)
#endif
#ifdef HAS_NV12TORGB24ROW_NEON
ANY21C(NV12ToRGB24Row_Any_NEON, NV12ToRGB24Row_NEON, 1, 1, 2, 3, 7)
#endif
#ifdef HAS_NV21TORGB24ROW_NEON
ANY21C(NV21ToRGB24Row_Any_NEON, NV21ToRGB24Row_NEON, 1, 1, 2, 3, 7)
#endif
#ifdef HAS_NV12TORGB24ROW_SSSE3
ANY21C(NV12ToRGB24Row_Any_SSSE3, NV12ToRGB24Row_SSSE3, 1, 1, 2, 3, 15)
#endif
#ifdef HAS_NV12TORGB24ROW_MMI
ANY21C(NV12ToRGB24Row_Any_MMI, NV12ToRGB24Row_MMI, 1, 1, 2, 3, 7)
#endif
#ifdef HAS_NV21TORGB24ROW_SSSE3
ANY21C(NV21ToRGB24Row_Any_SSSE3, NV21ToRGB24Row_SSSE3, 1, 1, 2, 3, 15)
#endif
#ifdef HAS_NV12TORGB24ROW_AVX2
ANY21C(NV12ToRGB24Row_Any_AVX2, NV12ToRGB24Row_AVX2, 1, 1, 2, 3, 31)
#endif
#ifdef HAS_NV21TORGB24ROW_AVX2
ANY21C(NV21ToRGB24Row_Any_AVX2, NV21ToRGB24Row_AVX2, 1, 1, 2, 3, 31)
#endif
#ifdef HAS_NV21TORGB24ROW_MMI
ANY21C(NV21ToRGB24Row_Any_MMI, NV21ToRGB24Row_MMI, 1, 1, 2, 3, 7)
#endif
#ifdef HAS_NV12TORGB565ROW_SSSE3
ANY21C(NV12ToRGB565Row_Any_SSSE3, NV12ToRGB565Row_SSSE3, 1, 1, 2, 2, 7)
#endif
#ifdef HAS_NV12TORGB565ROW_AVX2
ANY21C(NV12ToRGB565Row_Any_AVX2, NV12ToRGB565Row_AVX2, 1, 1, 2, 2, 15)
#endif
#ifdef HAS_NV12TORGB565ROW_NEON
ANY21C(NV12ToRGB565Row_Any_NEON, NV12ToRGB565Row_NEON, 1, 1, 2, 2, 7)
#endif
#ifdef HAS_NV12TORGB565ROW_MSA
ANY21C(NV12ToRGB565Row_Any_MSA, NV12ToRGB565Row_MSA, 1, 1, 2, 2, 7)
#endif
#ifdef HAS_NV12TORGB565ROW_MMI
ANY21C(NV12ToRGB565Row_Any_MMI, NV12ToRGB565Row_MMI, 1, 1, 2, 2, 7)
#endif
#undef ANY21C
// Any 1 to 1.
#define ANY11(NAMEANY, ANY_SIMD, UVSHIFT, SBPP, BPP, MASK) \
void NAMEANY(const uint8_t* src_ptr, uint8_t* dst_ptr, int width) { \
SIMD_ALIGNED(uint8_t temp[128 * 2]); \
memset(temp, 0, 128); /* for YUY2 and msan */ \
int r = width & MASK; \
int n = width & ~MASK; \
if (n > 0) { \
ANY_SIMD(src_ptr, dst_ptr, n); \
} \
memcpy(temp, src_ptr + (n >> UVSHIFT) * SBPP, SS(r, UVSHIFT) * SBPP); \
ANY_SIMD(temp, temp + 128, MASK + 1); \
memcpy(dst_ptr + n * BPP, temp + 128, r * BPP); \
}
#ifdef HAS_COPYROW_AVX
ANY11(CopyRow_Any_AVX, CopyRow_AVX, 0, 1, 1, 63)
#endif
#ifdef HAS_COPYROW_SSE2
ANY11(CopyRow_Any_SSE2, CopyRow_SSE2, 0, 1, 1, 31)
#endif
#ifdef HAS_COPYROW_NEON
ANY11(CopyRow_Any_NEON, CopyRow_NEON, 0, 1, 1, 31)
#endif
#if defined(HAS_ARGBTORGB24ROW_SSSE3)
ANY11(ARGBToRGB24Row_Any_SSSE3, ARGBToRGB24Row_SSSE3, 0, 4, 3, 15)
ANY11(ARGBToRAWRow_Any_SSSE3, ARGBToRAWRow_SSSE3, 0, 4, 3, 15)
ANY11(ARGBToRGB565Row_Any_SSE2, ARGBToRGB565Row_SSE2, 0, 4, 2, 3)
ANY11(ARGBToARGB1555Row_Any_SSE2, ARGBToARGB1555Row_SSE2, 0, 4, 2, 3)
ANY11(ARGBToARGB4444Row_Any_SSE2, ARGBToARGB4444Row_SSE2, 0, 4, 2, 3)
#endif
#if defined(HAS_ARGBTORGB24ROW_AVX2)
ANY11(ARGBToRGB24Row_Any_AVX2, ARGBToRGB24Row_AVX2, 0, 4, 3, 31)
#endif
#if defined(HAS_ARGBTORGB24ROW_AVX512VBMI)
ANY11(ARGBToRGB24Row_Any_AVX512VBMI, ARGBToRGB24Row_AVX512VBMI, 0, 4, 3, 31)
#endif
#if defined(HAS_ARGBTORAWROW_AVX2)
ANY11(ARGBToRAWRow_Any_AVX2, ARGBToRAWRow_AVX2, 0, 4, 3, 31)
#endif
#if defined(HAS_ARGBTORGB565ROW_AVX2)
ANY11(ARGBToRGB565Row_Any_AVX2, ARGBToRGB565Row_AVX2, 0, 4, 2, 7)
#endif
#if defined(HAS_ARGBTOARGB4444ROW_AVX2)
ANY11(ARGBToARGB1555Row_Any_AVX2, ARGBToARGB1555Row_AVX2, 0, 4, 2, 7)
ANY11(ARGBToARGB4444Row_Any_AVX2, ARGBToARGB4444Row_AVX2, 0, 4, 2, 7)
#endif
#if defined(HAS_ABGRTOAR30ROW_SSSE3)
ANY11(ABGRToAR30Row_Any_SSSE3, ABGRToAR30Row_SSSE3, 0, 4, 4, 3)
#endif
#if defined(HAS_ARGBTOAR30ROW_SSSE3)
ANY11(ARGBToAR30Row_Any_SSSE3, ARGBToAR30Row_SSSE3, 0, 4, 4, 3)
#endif
#if defined(HAS_ABGRTOAR30ROW_AVX2)
ANY11(ABGRToAR30Row_Any_AVX2, ABGRToAR30Row_AVX2, 0, 4, 4, 7)
#endif
#if defined(HAS_ARGBTOAR30ROW_AVX2)
ANY11(ARGBToAR30Row_Any_AVX2, ARGBToAR30Row_AVX2, 0, 4, 4, 7)
#endif
#if defined(HAS_J400TOARGBROW_SSE2)
ANY11(J400ToARGBRow_Any_SSE2, J400ToARGBRow_SSE2, 0, 1, 4, 7)
#endif
#if defined(HAS_J400TOARGBROW_AVX2)
ANY11(J400ToARGBRow_Any_AVX2, J400ToARGBRow_AVX2, 0, 1, 4, 15)
#endif
#if defined(HAS_RGB24TOARGBROW_SSSE3)
ANY11(RGB24ToARGBRow_Any_SSSE3, RGB24ToARGBRow_SSSE3, 0, 3, 4, 15)
ANY11(RAWToARGBRow_Any_SSSE3, RAWToARGBRow_SSSE3, 0, 3, 4, 15)
ANY11(RGB565ToARGBRow_Any_SSE2, RGB565ToARGBRow_SSE2, 0, 2, 4, 7)
ANY11(ARGB1555ToARGBRow_Any_SSE2, ARGB1555ToARGBRow_SSE2, 0, 2, 4, 7)
ANY11(ARGB4444ToARGBRow_Any_SSE2, ARGB4444ToARGBRow_SSE2, 0, 2, 4, 7)
#endif
#if defined(HAS_RAWTORGBAROW_SSSE3)
ANY11(RAWToRGBARow_Any_SSSE3, RAWToRGBARow_SSSE3, 0, 3, 4, 15)
#endif
#if defined(HAS_RAWTORGB24ROW_SSSE3)
ANY11(RAWToRGB24Row_Any_SSSE3, RAWToRGB24Row_SSSE3, 0, 3, 3, 7)
#endif
#if defined(HAS_RGB565TOARGBROW_AVX2)
ANY11(RGB565ToARGBRow_Any_AVX2, RGB565ToARGBRow_AVX2, 0, 2, 4, 15)
#endif
#if defined(HAS_ARGB1555TOARGBROW_AVX2)
ANY11(ARGB1555ToARGBRow_Any_AVX2, ARGB1555ToARGBRow_AVX2, 0, 2, 4, 15)
#endif
#if defined(HAS_ARGB4444TOARGBROW_AVX2)
ANY11(ARGB4444ToARGBRow_Any_AVX2, ARGB4444ToARGBRow_AVX2, 0, 2, 4, 15)
#endif
#if defined(HAS_ARGBTORGB24ROW_NEON)
ANY11(ARGBToRGB24Row_Any_NEON, ARGBToRGB24Row_NEON, 0, 4, 3, 7)
ANY11(ARGBToRAWRow_Any_NEON, ARGBToRAWRow_NEON, 0, 4, 3, 7)
ANY11(ARGBToRGB565Row_Any_NEON, ARGBToRGB565Row_NEON, 0, 4, 2, 7)
ANY11(ARGBToARGB1555Row_Any_NEON, ARGBToARGB1555Row_NEON, 0, 4, 2, 7)
ANY11(ARGBToARGB4444Row_Any_NEON, ARGBToARGB4444Row_NEON, 0, 4, 2, 7)
ANY11(J400ToARGBRow_Any_NEON, J400ToARGBRow_NEON, 0, 1, 4, 7)
#endif
#if defined(HAS_ARGBTORGB24ROW_MSA)
ANY11(ARGBToRGB24Row_Any_MSA, ARGBToRGB24Row_MSA, 0, 4, 3, 15)
ANY11(ARGBToRAWRow_Any_MSA, ARGBToRAWRow_MSA, 0, 4, 3, 15)
ANY11(ARGBToRGB565Row_Any_MSA, ARGBToRGB565Row_MSA, 0, 4, 2, 7)
ANY11(ARGBToARGB1555Row_Any_MSA, ARGBToARGB1555Row_MSA, 0, 4, 2, 7)
ANY11(ARGBToARGB4444Row_Any_MSA, ARGBToARGB4444Row_MSA, 0, 4, 2, 7)
ANY11(J400ToARGBRow_Any_MSA, J400ToARGBRow_MSA, 0, 1, 4, 15)
#endif
#if defined(HAS_ARGBTORGB24ROW_MMI)
ANY11(ARGBToRGB24Row_Any_MMI, ARGBToRGB24Row_MMI, 0, 4, 3, 3)
ANY11(ARGBToRAWRow_Any_MMI, ARGBToRAWRow_MMI, 0, 4, 3, 3)
ANY11(ARGBToRGB565Row_Any_MMI, ARGBToRGB565Row_MMI, 0, 4, 2, 3)
ANY11(ARGBToARGB1555Row_Any_MMI, ARGBToARGB1555Row_MMI, 0, 4, 2, 3)
ANY11(ARGBToARGB4444Row_Any_MMI, ARGBToARGB4444Row_MMI, 0, 4, 2, 3)
ANY11(J400ToARGBRow_Any_MMI, J400ToARGBRow_MMI, 0, 1, 4, 3)
#endif
#if defined(HAS_RAWTORGB24ROW_NEON)
ANY11(RAWToRGB24Row_Any_NEON, RAWToRGB24Row_NEON, 0, 3, 3, 7)
#endif
#if defined(HAS_RAWTORGB24ROW_MSA)
ANY11(RAWToRGB24Row_Any_MSA, RAWToRGB24Row_MSA, 0, 3, 3, 15)
#endif
#if defined(HAS_RAWTORGB24ROW_MMI)
ANY11(RAWToRGB24Row_Any_MMI, RAWToRGB24Row_MMI, 0, 3, 3, 3)
#endif
#ifdef HAS_ARGBTOYROW_AVX2
ANY11(ARGBToYRow_Any_AVX2, ARGBToYRow_AVX2, 0, 4, 1, 31)
#endif
#ifdef HAS_ABGRTOYROW_AVX2
ANY11(ABGRToYRow_Any_AVX2, ABGRToYRow_AVX2, 0, 4, 1, 31)
#endif
#ifdef HAS_ARGBTOYJROW_AVX2
ANY11(ARGBToYJRow_Any_AVX2, ARGBToYJRow_AVX2, 0, 4, 1, 31)
#endif
#ifdef HAS_RGBATOYJROW_AVX2
ANY11(RGBAToYJRow_Any_AVX2, RGBAToYJRow_AVX2, 0, 4, 1, 31)
#endif
#ifdef HAS_UYVYTOYROW_AVX2
ANY11(UYVYToYRow_Any_AVX2, UYVYToYRow_AVX2, 0, 2, 1, 31)
#endif
#ifdef HAS_YUY2TOYROW_AVX2
ANY11(YUY2ToYRow_Any_AVX2, YUY2ToYRow_AVX2, 1, 4, 1, 31)
#endif
#ifdef HAS_ARGBTOYROW_SSSE3
ANY11(ARGBToYRow_Any_SSSE3, ARGBToYRow_SSSE3, 0, 4, 1, 15)
#endif
#ifdef HAS_BGRATOYROW_SSSE3
ANY11(BGRAToYRow_Any_SSSE3, BGRAToYRow_SSSE3, 0, 4, 1, 15)
ANY11(ABGRToYRow_Any_SSSE3, ABGRToYRow_SSSE3, 0, 4, 1, 15)
ANY11(RGBAToYRow_Any_SSSE3, RGBAToYRow_SSSE3, 0, 4, 1, 15)
ANY11(YUY2ToYRow_Any_SSE2, YUY2ToYRow_SSE2, 1, 4, 1, 15)
ANY11(UYVYToYRow_Any_SSE2, UYVYToYRow_SSE2, 1, 4, 1, 15)
#endif
#ifdef HAS_ARGBTOYJROW_SSSE3
ANY11(ARGBToYJRow_Any_SSSE3, ARGBToYJRow_SSSE3, 0, 4, 1, 15)
#endif
#ifdef HAS_RGBATOYJROW_SSSE3
ANY11(RGBAToYJRow_Any_SSSE3, RGBAToYJRow_SSSE3, 0, 4, 1, 15)
#endif
#ifdef HAS_ARGBTOYROW_NEON
ANY11(ARGBToYRow_Any_NEON, ARGBToYRow_NEON, 0, 4, 1, 7)
#endif
#ifdef HAS_ARGBTOYROW_MSA
ANY11(ARGBToYRow_Any_MSA, ARGBToYRow_MSA, 0, 4, 1, 15)
#endif
#ifdef HAS_ARGBTOYROW_MMI
ANY11(ARGBToYRow_Any_MMI, ARGBToYRow_MMI, 0, 4, 1, 7)
#endif
#ifdef HAS_ARGBTOYJROW_NEON
ANY11(ARGBToYJRow_Any_NEON, ARGBToYJRow_NEON, 0, 4, 1, 7)
#endif
#ifdef HAS_RGBATOYJROW_NEON
ANY11(RGBAToYJRow_Any_NEON, RGBAToYJRow_NEON, 0, 4, 1, 7)
#endif
#ifdef HAS_ARGBTOYJROW_MSA
ANY11(ARGBToYJRow_Any_MSA, ARGBToYJRow_MSA, 0, 4, 1, 15)
#endif
#ifdef HAS_ARGBTOYJROW_MMI
ANY11(ARGBToYJRow_Any_MMI, ARGBToYJRow_MMI, 0, 4, 1, 7)
#endif
#ifdef HAS_BGRATOYROW_NEON
ANY11(BGRAToYRow_Any_NEON, BGRAToYRow_NEON, 0, 4, 1, 7)
#endif
#ifdef HAS_BGRATOYROW_MSA
ANY11(BGRAToYRow_Any_MSA, BGRAToYRow_MSA, 0, 4, 1, 15)
#endif
#ifdef HAS_BGRATOYROW_MMI
ANY11(BGRAToYRow_Any_MMI, BGRAToYRow_MMI, 0, 4, 1, 7)
#endif
#ifdef HAS_ABGRTOYROW_NEON
ANY11(ABGRToYRow_Any_NEON, ABGRToYRow_NEON, 0, 4, 1, 7)
#endif
#ifdef HAS_ABGRTOYROW_MSA
ANY11(ABGRToYRow_Any_MSA, ABGRToYRow_MSA, 0, 4, 1, 7)
#endif
#ifdef HAS_ABGRTOYROW_MMI
ANY11(ABGRToYRow_Any_MMI, ABGRToYRow_MMI, 0, 4, 1, 7)
#endif
#ifdef HAS_RGBATOYROW_NEON
ANY11(RGBAToYRow_Any_NEON, RGBAToYRow_NEON, 0, 4, 1, 7)
#endif
#ifdef HAS_RGBATOYROW_MSA
ANY11(RGBAToYRow_Any_MSA, RGBAToYRow_MSA, 0, 4, 1, 15)
#endif
#ifdef HAS_RGBATOYROW_MMI
ANY11(RGBAToYRow_Any_MMI, RGBAToYRow_MMI, 0, 4, 1, 7)
#endif
#ifdef HAS_RGB24TOYROW_NEON
ANY11(RGB24ToYRow_Any_NEON, RGB24ToYRow_NEON, 0, 3, 1, 7)
#endif
#ifdef HAS_RGB24TOYJROW_AVX2
ANY11(RGB24ToYJRow_Any_AVX2, RGB24ToYJRow_AVX2, 0, 3, 1, 31)
#endif
#ifdef HAS_RGB24TOYJROW_SSSE3
ANY11(RGB24ToYJRow_Any_SSSE3, RGB24ToYJRow_SSSE3, 0, 3, 1, 15)
#endif
#ifdef HAS_RGB24TOYJROW_NEON
ANY11(RGB24ToYJRow_Any_NEON, RGB24ToYJRow_NEON, 0, 3, 1, 7)
#endif
#ifdef HAS_RGB24TOYROW_MSA
ANY11(RGB24ToYRow_Any_MSA, RGB24ToYRow_MSA, 0, 3, 1, 15)
#endif
#ifdef HAS_RGB24TOYROW_MMI
ANY11(RGB24ToYRow_Any_MMI, RGB24ToYRow_MMI, 0, 3, 1, 7)
#endif
#ifdef HAS_RAWTOYROW_NEON
ANY11(RAWToYRow_Any_NEON, RAWToYRow_NEON, 0, 3, 1, 7)
#endif
#ifdef HAS_RAWTOYJROW_AVX2
ANY11(RAWToYJRow_Any_AVX2, RAWToYJRow_AVX2, 0, 3, 1, 31)
#endif
#ifdef HAS_RAWTOYJROW_SSSE3
ANY11(RAWToYJRow_Any_SSSE3, RAWToYJRow_SSSE3, 0, 3, 1, 15)
#endif
#ifdef HAS_RAWTOYJROW_NEON
ANY11(RAWToYJRow_Any_NEON, RAWToYJRow_NEON, 0, 3, 1, 7)
#endif
#ifdef HAS_RAWTOYROW_MSA
ANY11(RAWToYRow_Any_MSA, RAWToYRow_MSA, 0, 3, 1, 15)
#endif
#ifdef HAS_RAWTOYROW_MMI
ANY11(RAWToYRow_Any_MMI, RAWToYRow_MMI, 0, 3, 1, 7)
#endif
#ifdef HAS_RGB565TOYROW_NEON
ANY11(RGB565ToYRow_Any_NEON, RGB565ToYRow_NEON, 0, 2, 1, 7)
#endif
#ifdef HAS_RGB565TOYROW_MSA
ANY11(RGB565ToYRow_Any_MSA, RGB565ToYRow_MSA, 0, 2, 1, 15)
#endif
#ifdef HAS_RGB565TOYROW_MMI
ANY11(RGB565ToYRow_Any_MMI, RGB565ToYRow_MMI, 0, 2, 1, 7)
#endif
#ifdef HAS_ARGB1555TOYROW_NEON
ANY11(ARGB1555ToYRow_Any_NEON, ARGB1555ToYRow_NEON, 0, 2, 1, 7)
#endif
#ifdef HAS_ARGB1555TOYROW_MSA
ANY11(ARGB1555ToYRow_Any_MSA, ARGB1555ToYRow_MSA, 0, 2, 1, 15)
#endif
#ifdef HAS_ARGB1555TOYROW_MMI
ANY11(ARGB1555ToYRow_Any_MMI, ARGB1555ToYRow_MMI, 0, 2, 1, 7)
#endif
#ifdef HAS_ARGB4444TOYROW_NEON
ANY11(ARGB4444ToYRow_Any_NEON, ARGB4444ToYRow_NEON, 0, 2, 1, 7)
#endif
#ifdef HAS_ARGB4444TOYROW_MMI
ANY11(ARGB4444ToYRow_Any_MMI, ARGB4444ToYRow_MMI, 0, 2, 1, 7)
#endif
#ifdef HAS_YUY2TOYROW_NEON
ANY11(YUY2ToYRow_Any_NEON, YUY2ToYRow_NEON, 1, 4, 1, 15)
#endif
#ifdef HAS_UYVYTOYROW_NEON
ANY11(UYVYToYRow_Any_NEON, UYVYToYRow_NEON, 1, 4, 1, 15)
#endif
#ifdef HAS_YUY2TOYROW_MSA
ANY11(YUY2ToYRow_Any_MSA, YUY2ToYRow_MSA, 1, 4, 1, 31)
#endif
#ifdef HAS_YUY2TOYROW_MMI
ANY11(YUY2ToYRow_Any_MMI, YUY2ToYRow_MMI, 1, 4, 1, 7)
#endif
#ifdef HAS_UYVYTOYROW_MSA
ANY11(UYVYToYRow_Any_MSA, UYVYToYRow_MSA, 1, 4, 1, 31)
#endif
#ifdef HAS_UYVYTOYROW_MMI
ANY11(UYVYToYRow_Any_MMI, UYVYToYRow_MMI, 1, 4, 1, 15)
#endif
#ifdef HAS_AYUVTOYROW_NEON
ANY11(AYUVToYRow_Any_NEON, AYUVToYRow_NEON, 0, 4, 1, 15)
#endif
#ifdef HAS_SWAPUVROW_SSSE3
ANY11(SwapUVRow_Any_SSSE3, SwapUVRow_SSSE3, 0, 2, 2, 15)
#endif
#ifdef HAS_SWAPUVROW_AVX2
ANY11(SwapUVRow_Any_AVX2, SwapUVRow_AVX2, 0, 2, 2, 31)
#endif
#ifdef HAS_SWAPUVROW_NEON
ANY11(SwapUVRow_Any_NEON, SwapUVRow_NEON, 0, 2, 2, 15)
#endif
#ifdef HAS_RGB24TOARGBROW_NEON
ANY11(RGB24ToARGBRow_Any_NEON, RGB24ToARGBRow_NEON, 0, 3, 4, 7)
#endif
#ifdef HAS_RGB24TOARGBROW_MSA
ANY11(RGB24ToARGBRow_Any_MSA, RGB24ToARGBRow_MSA, 0, 3, 4, 15)
#endif
#ifdef HAS_RGB24TOARGBROW_MMI
ANY11(RGB24ToARGBRow_Any_MMI, RGB24ToARGBRow_MMI, 0, 3, 4, 3)
#endif
#ifdef HAS_RAWTOARGBROW_NEON
ANY11(RAWToARGBRow_Any_NEON, RAWToARGBRow_NEON, 0, 3, 4, 7)
#endif
#ifdef HAS_RAWTORGBAROW_NEON
ANY11(RAWToRGBARow_Any_NEON, RAWToRGBARow_NEON, 0, 3, 4, 7)
#endif
#ifdef HAS_RAWTOARGBROW_MSA
ANY11(RAWToARGBRow_Any_MSA, RAWToARGBRow_MSA, 0, 3, 4, 15)
#endif
#ifdef HAS_RAWTOARGBROW_MMI
ANY11(RAWToARGBRow_Any_MMI, RAWToARGBRow_MMI, 0, 3, 4, 3)
#endif
#ifdef HAS_RGB565TOARGBROW_NEON
ANY11(RGB565ToARGBRow_Any_NEON, RGB565ToARGBRow_NEON, 0, 2, 4, 7)
#endif
#ifdef HAS_RGB565TOARGBROW_MSA
ANY11(RGB565ToARGBRow_Any_MSA, RGB565ToARGBRow_MSA, 0, 2, 4, 15)
#endif
#ifdef HAS_RGB565TOARGBROW_MMI
ANY11(RGB565ToARGBRow_Any_MMI, RGB565ToARGBRow_MMI, 0, 2, 4, 3)
#endif
#ifdef HAS_ARGB1555TOARGBROW_NEON
ANY11(ARGB1555ToARGBRow_Any_NEON, ARGB1555ToARGBRow_NEON, 0, 2, 4, 7)
#endif
#ifdef HAS_ARGB1555TOARGBROW_MSA
ANY11(ARGB1555ToARGBRow_Any_MSA, ARGB1555ToARGBRow_MSA, 0, 2, 4, 15)
#endif
#ifdef HAS_ARGB1555TOARGBROW_MMI
ANY11(ARGB1555ToARGBRow_Any_MMI, ARGB1555ToARGBRow_MMI, 0, 2, 4, 3)
#endif
#ifdef HAS_ARGB4444TOARGBROW_NEON
ANY11(ARGB4444ToARGBRow_Any_NEON, ARGB4444ToARGBRow_NEON, 0, 2, 4, 7)
#endif
#ifdef HAS_ARGB4444TOARGBROW_MSA
ANY11(ARGB4444ToARGBRow_Any_MSA, ARGB4444ToARGBRow_MSA, 0, 2, 4, 15)
#endif
#ifdef HAS_ARGB4444TOARGBROW_MMI
ANY11(ARGB4444ToARGBRow_Any_MMI, ARGB4444ToARGBRow_MMI, 0, 2, 4, 3)
#endif
#ifdef HAS_ARGBATTENUATEROW_SSSE3
ANY11(ARGBAttenuateRow_Any_SSSE3, ARGBAttenuateRow_SSSE3, 0, 4, 4, 3)
#endif
#ifdef HAS_ARGBUNATTENUATEROW_SSE2
ANY11(ARGBUnattenuateRow_Any_SSE2, ARGBUnattenuateRow_SSE2, 0, 4, 4, 3)
#endif
#ifdef HAS_ARGBATTENUATEROW_AVX2
ANY11(ARGBAttenuateRow_Any_AVX2, ARGBAttenuateRow_AVX2, 0, 4, 4, 7)
#endif
#ifdef HAS_ARGBUNATTENUATEROW_AVX2
ANY11(ARGBUnattenuateRow_Any_AVX2, ARGBUnattenuateRow_AVX2, 0, 4, 4, 7)
#endif
#ifdef HAS_ARGBATTENUATEROW_NEON
ANY11(ARGBAttenuateRow_Any_NEON, ARGBAttenuateRow_NEON, 0, 4, 4, 7)
#endif
#ifdef HAS_ARGBATTENUATEROW_MSA
ANY11(ARGBAttenuateRow_Any_MSA, ARGBAttenuateRow_MSA, 0, 4, 4, 7)
#endif
#ifdef HAS_ARGBATTENUATEROW_MMI
ANY11(ARGBAttenuateRow_Any_MMI, ARGBAttenuateRow_MMI, 0, 4, 4, 1)
#endif
#ifdef HAS_ARGBEXTRACTALPHAROW_SSE2
ANY11(ARGBExtractAlphaRow_Any_SSE2, ARGBExtractAlphaRow_SSE2, 0, 4, 1, 7)
#endif
#ifdef HAS_ARGBEXTRACTALPHAROW_AVX2
ANY11(ARGBExtractAlphaRow_Any_AVX2, ARGBExtractAlphaRow_AVX2, 0, 4, 1, 31)
#endif
#ifdef HAS_ARGBEXTRACTALPHAROW_NEON
ANY11(ARGBExtractAlphaRow_Any_NEON, ARGBExtractAlphaRow_NEON, 0, 4, 1, 15)
#endif
#ifdef HAS_ARGBEXTRACTALPHAROW_MSA
ANY11(ARGBExtractAlphaRow_Any_MSA, ARGBExtractAlphaRow_MSA, 0, 4, 1, 15)
#endif
#ifdef HAS_ARGBEXTRACTALPHAROW_MMI
ANY11(ARGBExtractAlphaRow_Any_MMI, ARGBExtractAlphaRow_MMI, 0, 4, 1, 7)
#endif
#undef ANY11
// Any 1 to 1 blended. Destination is read, modify, write.
#define ANY11B(NAMEANY, ANY_SIMD, UVSHIFT, SBPP, BPP, MASK) \
void NAMEANY(const uint8_t* src_ptr, uint8_t* dst_ptr, int width) { \
SIMD_ALIGNED(uint8_t temp[64 * 2]); \
memset(temp, 0, 64 * 2); /* for msan */ \
int r = width & MASK; \
int n = width & ~MASK; \
if (n > 0) { \
ANY_SIMD(src_ptr, dst_ptr, n); \
} \
memcpy(temp, src_ptr + (n >> UVSHIFT) * SBPP, SS(r, UVSHIFT) * SBPP); \
memcpy(temp + 64, dst_ptr + n * BPP, r * BPP); \
ANY_SIMD(temp, temp + 64, MASK + 1); \
memcpy(dst_ptr + n * BPP, temp + 64, r * BPP); \
}
#ifdef HAS_ARGBCOPYALPHAROW_AVX2
ANY11B(ARGBCopyAlphaRow_Any_AVX2, ARGBCopyAlphaRow_AVX2, 0, 4, 4, 15)
#endif
#ifdef HAS_ARGBCOPYALPHAROW_SSE2
ANY11B(ARGBCopyAlphaRow_Any_SSE2, ARGBCopyAlphaRow_SSE2, 0, 4, 4, 7)
#endif
#ifdef HAS_ARGBCOPYALPHAROW_MMI
ANY11B(ARGBCopyAlphaRow_Any_MMI, ARGBCopyAlphaRow_MMI, 0, 4, 4, 1)
#endif
#ifdef HAS_ARGBCOPYYTOALPHAROW_AVX2
ANY11B(ARGBCopyYToAlphaRow_Any_AVX2, ARGBCopyYToAlphaRow_AVX2, 0, 1, 4, 15)
#endif
#ifdef HAS_ARGBCOPYYTOALPHAROW_SSE2
ANY11B(ARGBCopyYToAlphaRow_Any_SSE2, ARGBCopyYToAlphaRow_SSE2, 0, 1, 4, 7)
#endif
#ifdef HAS_ARGBCOPYYTOALPHAROW_MMI
ANY11B(ARGBCopyYToAlphaRow_Any_MMI, ARGBCopyYToAlphaRow_MMI, 0, 1, 4, 7)
#endif
#undef ANY11B
// Any 1 to 1 with parameter.
#define ANY11P(NAMEANY, ANY_SIMD, T, SBPP, BPP, MASK) \
void NAMEANY(const uint8_t* src_ptr, uint8_t* dst_ptr, T param, int width) { \
SIMD_ALIGNED(uint8_t temp[64 * 2]); \
memset(temp, 0, 64); /* for msan */ \
int r = width & MASK; \
int n = width & ~MASK; \
if (n > 0) { \
ANY_SIMD(src_ptr, dst_ptr, param, n); \
} \
memcpy(temp, src_ptr + n * SBPP, r * SBPP); \
ANY_SIMD(temp, temp + 64, param, MASK + 1); \
memcpy(dst_ptr + n * BPP, temp + 64, r * BPP); \
}
#if defined(HAS_I400TOARGBROW_SSE2)
ANY11P(I400ToARGBRow_Any_SSE2,
I400ToARGBRow_SSE2,
const struct YuvConstants*,
1,
4,
7)
#endif
#if defined(HAS_I400TOARGBROW_AVX2)
ANY11P(I400ToARGBRow_Any_AVX2,
I400ToARGBRow_AVX2,
const struct YuvConstants*,
1,
4,
15)
#endif
#if defined(HAS_I400TOARGBROW_NEON)
ANY11P(I400ToARGBRow_Any_NEON,
I400ToARGBRow_NEON,
const struct YuvConstants*,
1,
4,
7)
#endif
#if defined(HAS_I400TOARGBROW_MSA)
ANY11P(I400ToARGBRow_Any_MSA,
I400ToARGBRow_MSA,
const struct YuvConstants*,
1,
4,
15)
#endif
#if defined(HAS_I400TOARGBROW_MMI)
ANY11P(I400ToARGBRow_Any_MMI,
I400ToARGBRow_MMI,
const struct YuvConstants*,
1,
4,
7)
#endif
#if defined(HAS_ARGBTORGB565DITHERROW_SSE2)
ANY11P(ARGBToRGB565DitherRow_Any_SSE2,
ARGBToRGB565DitherRow_SSE2,
const uint32_t,
4,
2,
3)
#endif
#if defined(HAS_ARGBTORGB565DITHERROW_AVX2)
ANY11P(ARGBToRGB565DitherRow_Any_AVX2,
ARGBToRGB565DitherRow_AVX2,
const uint32_t,
4,
2,
7)
#endif
#if defined(HAS_ARGBTORGB565DITHERROW_NEON)
ANY11P(ARGBToRGB565DitherRow_Any_NEON,
ARGBToRGB565DitherRow_NEON,
const uint32_t,
4,
2,
7)
#endif
#if defined(HAS_ARGBTORGB565DITHERROW_MSA)
ANY11P(ARGBToRGB565DitherRow_Any_MSA,
ARGBToRGB565DitherRow_MSA,
const uint32_t,
4,
2,
7)
#endif
#if defined(HAS_ARGBTORGB565DITHERROW_MMI)
ANY11P(ARGBToRGB565DitherRow_Any_MMI,
ARGBToRGB565DitherRow_MMI,
const uint32_t,
4,
2,
3)
#endif
#ifdef HAS_ARGBSHUFFLEROW_SSSE3
ANY11P(ARGBShuffleRow_Any_SSSE3, ARGBShuffleRow_SSSE3, const uint8_t*, 4, 4, 7)
#endif
#ifdef HAS_ARGBSHUFFLEROW_AVX2
ANY11P(ARGBShuffleRow_Any_AVX2, ARGBShuffleRow_AVX2, const uint8_t*, 4, 4, 15)
#endif
#ifdef HAS_ARGBSHUFFLEROW_NEON
ANY11P(ARGBShuffleRow_Any_NEON, ARGBShuffleRow_NEON, const uint8_t*, 4, 4, 3)
#endif
#ifdef HAS_ARGBSHUFFLEROW_MSA
ANY11P(ARGBShuffleRow_Any_MSA, ARGBShuffleRow_MSA, const uint8_t*, 4, 4, 7)
#endif
#ifdef HAS_ARGBSHUFFLEROW_MMI
ANY11P(ARGBShuffleRow_Any_MMI, ARGBShuffleRow_MMI, const uint8_t*, 4, 4, 1)
#endif
#undef ANY11P
#undef ANY11P
// Any 1 to 1 with parameter and shorts. BPP measures in shorts.
#define ANY11C(NAMEANY, ANY_SIMD, SBPP, BPP, STYPE, DTYPE, MASK) \
void NAMEANY(const STYPE* src_ptr, DTYPE* dst_ptr, int scale, int width) { \
SIMD_ALIGNED(STYPE temp[32]); \
SIMD_ALIGNED(DTYPE out[32]); \
memset(temp, 0, 32 * SBPP); /* for msan */ \
int r = width & MASK; \
int n = width & ~MASK; \
if (n > 0) { \
ANY_SIMD(src_ptr, dst_ptr, scale, n); \
} \
memcpy(temp, src_ptr + n, r * SBPP); \
ANY_SIMD(temp, out, scale, MASK + 1); \
memcpy(dst_ptr + n, out, r * BPP); \
}
#ifdef HAS_CONVERT16TO8ROW_SSSE3
ANY11C(Convert16To8Row_Any_SSSE3,
Convert16To8Row_SSSE3,
2,
1,
uint16_t,
uint8_t,
15)
#endif
#ifdef HAS_CONVERT16TO8ROW_AVX2
ANY11C(Convert16To8Row_Any_AVX2,
Convert16To8Row_AVX2,
2,
1,
uint16_t,
uint8_t,
31)
#endif
#ifdef HAS_CONVERT8TO16ROW_SSE2
ANY11C(Convert8To16Row_Any_SSE2,
Convert8To16Row_SSE2,
1,
2,
uint8_t,
uint16_t,
15)
#endif
#ifdef HAS_CONVERT8TO16ROW_AVX2
ANY11C(Convert8To16Row_Any_AVX2,
Convert8To16Row_AVX2,
1,
2,
uint8_t,
uint16_t,
31)
#endif
#undef ANY11C
// Any 1 to 1 with parameter and shorts to byte. BPP measures in shorts.
#define ANY11P16(NAMEANY, ANY_SIMD, ST, T, SBPP, BPP, MASK) \
void NAMEANY(const ST* src_ptr, T* dst_ptr, float param, int width) { \
SIMD_ALIGNED(ST temp[32]); \
SIMD_ALIGNED(T out[32]); \
memset(temp, 0, SBPP * 32); /* for msan */ \
int r = width & MASK; \
int n = width & ~MASK; \
if (n > 0) { \
ANY_SIMD(src_ptr, dst_ptr, param, n); \
} \
memcpy(temp, src_ptr + n, r * SBPP); \
ANY_SIMD(temp, out, param, MASK + 1); \
memcpy(dst_ptr + n, out, r * BPP); \
}
#ifdef HAS_HALFFLOATROW_SSE2
ANY11P16(HalfFloatRow_Any_SSE2, HalfFloatRow_SSE2, uint16_t, uint16_t, 2, 2, 7)
#endif
#ifdef HAS_HALFFLOATROW_AVX2
ANY11P16(HalfFloatRow_Any_AVX2, HalfFloatRow_AVX2, uint16_t, uint16_t, 2, 2, 15)
#endif
#ifdef HAS_HALFFLOATROW_F16C
ANY11P16(HalfFloatRow_Any_F16C, HalfFloatRow_F16C, uint16_t, uint16_t, 2, 2, 15)
ANY11P16(HalfFloat1Row_Any_F16C,
HalfFloat1Row_F16C,
uint16_t,
uint16_t,
2,
2,
15)
#endif
#ifdef HAS_HALFFLOATROW_NEON
ANY11P16(HalfFloatRow_Any_NEON, HalfFloatRow_NEON, uint16_t, uint16_t, 2, 2, 7)
ANY11P16(HalfFloat1Row_Any_NEON,
HalfFloat1Row_NEON,
uint16_t,
uint16_t,
2,
2,
7)
#endif
#ifdef HAS_HALFFLOATROW_MSA
ANY11P16(HalfFloatRow_Any_MSA, HalfFloatRow_MSA, uint16_t, uint16_t, 2, 2, 31)
#endif
#ifdef HAS_BYTETOFLOATROW_NEON
ANY11P16(ByteToFloatRow_Any_NEON, ByteToFloatRow_NEON, uint8_t, float, 1, 3, 7)
#endif
#undef ANY11P16
// Any 1 to 1 with yuvconstants
#define ANY11C(NAMEANY, ANY_SIMD, UVSHIFT, SBPP, BPP, MASK) \
void NAMEANY(const uint8_t* src_ptr, uint8_t* dst_ptr, \
const struct YuvConstants* yuvconstants, int width) { \
SIMD_ALIGNED(uint8_t temp[128 * 2]); \
memset(temp, 0, 128); /* for YUY2 and msan */ \
int r = width & MASK; \
int n = width & ~MASK; \
if (n > 0) { \
ANY_SIMD(src_ptr, dst_ptr, yuvconstants, n); \
} \
memcpy(temp, src_ptr + (n >> UVSHIFT) * SBPP, SS(r, UVSHIFT) * SBPP); \
ANY_SIMD(temp, temp + 128, yuvconstants, MASK + 1); \
memcpy(dst_ptr + n * BPP, temp + 128, r * BPP); \
}
#if defined(HAS_YUY2TOARGBROW_SSSE3)
ANY11C(YUY2ToARGBRow_Any_SSSE3, YUY2ToARGBRow_SSSE3, 1, 4, 4, 15)
ANY11C(UYVYToARGBRow_Any_SSSE3, UYVYToARGBRow_SSSE3, 1, 4, 4, 15)
#endif
#if defined(HAS_YUY2TOARGBROW_AVX2)
ANY11C(YUY2ToARGBRow_Any_AVX2, YUY2ToARGBRow_AVX2, 1, 4, 4, 31)
ANY11C(UYVYToARGBRow_Any_AVX2, UYVYToARGBRow_AVX2, 1, 4, 4, 31)
#endif
#if defined(HAS_YUY2TOARGBROW_NEON)
ANY11C(YUY2ToARGBRow_Any_NEON, YUY2ToARGBRow_NEON, 1, 4, 4, 7)
ANY11C(UYVYToARGBRow_Any_NEON, UYVYToARGBRow_NEON, 1, 4, 4, 7)
#endif
#if defined(HAS_YUY2TOARGBROW_MSA)
ANY11C(YUY2ToARGBRow_Any_MSA, YUY2ToARGBRow_MSA, 1, 4, 4, 7)
ANY11C(UYVYToARGBRow_Any_MSA, UYVYToARGBRow_MSA, 1, 4, 4, 7)
#endif
#if defined(HAS_YUY2TOARGBROW_MMI)
ANY11C(YUY2ToARGBRow_Any_MMI, YUY2ToARGBRow_MMI, 1, 4, 4, 7)
ANY11C(UYVYToARGBRow_Any_MMI, UYVYToARGBRow_MMI, 1, 4, 4, 7)
#endif
#undef ANY11C
// Any 1 to 1 interpolate. Takes 2 rows of source via stride.
#define ANY11T(NAMEANY, ANY_SIMD, SBPP, BPP, MASK) \
void NAMEANY(uint8_t* dst_ptr, const uint8_t* src_ptr, \
ptrdiff_t src_stride_ptr, int width, int source_y_fraction) { \
SIMD_ALIGNED(uint8_t temp[64 * 3]); \
memset(temp, 0, 64 * 2); /* for msan */ \
int r = width & MASK; \
int n = width & ~MASK; \
if (n > 0) { \
ANY_SIMD(dst_ptr, src_ptr, src_stride_ptr, n, source_y_fraction); \
} \
memcpy(temp, src_ptr + n * SBPP, r * SBPP); \
memcpy(temp + 64, src_ptr + src_stride_ptr + n * SBPP, r * SBPP); \
ANY_SIMD(temp + 128, temp, 64, MASK + 1, source_y_fraction); \
memcpy(dst_ptr + n * BPP, temp + 128, r * BPP); \
}
#ifdef HAS_INTERPOLATEROW_AVX2
ANY11T(InterpolateRow_Any_AVX2, InterpolateRow_AVX2, 1, 1, 31)
#endif
#ifdef HAS_INTERPOLATEROW_SSSE3
ANY11T(InterpolateRow_Any_SSSE3, InterpolateRow_SSSE3, 1, 1, 15)
#endif
#ifdef HAS_INTERPOLATEROW_NEON
ANY11T(InterpolateRow_Any_NEON, InterpolateRow_NEON, 1, 1, 15)
#endif
#ifdef HAS_INTERPOLATEROW_MSA
ANY11T(InterpolateRow_Any_MSA, InterpolateRow_MSA, 1, 1, 31)
#endif
#ifdef HAS_INTERPOLATEROW_MMI
ANY11T(InterpolateRow_Any_MMI, InterpolateRow_MMI, 1, 1, 7)
#endif
#undef ANY11T
// Any 1 to 1 mirror.
#define ANY11M(NAMEANY, ANY_SIMD, BPP, MASK) \
void NAMEANY(const uint8_t* src_ptr, uint8_t* dst_ptr, int width) { \
SIMD_ALIGNED(uint8_t temp[64 * 2]); \
memset(temp, 0, 64); /* for msan */ \
int r = width & MASK; \
int n = width & ~MASK; \
if (n > 0) { \
ANY_SIMD(src_ptr + r * BPP, dst_ptr, n); \
} \
memcpy(temp, src_ptr, r* BPP); \
ANY_SIMD(temp, temp + 64, MASK + 1); \
memcpy(dst_ptr + n * BPP, temp + 64 + (MASK + 1 - r) * BPP, r * BPP); \
}
#ifdef HAS_MIRRORROW_AVX2
ANY11M(MirrorRow_Any_AVX2, MirrorRow_AVX2, 1, 31)
#endif
#ifdef HAS_MIRRORROW_SSSE3
ANY11M(MirrorRow_Any_SSSE3, MirrorRow_SSSE3, 1, 15)
#endif
#ifdef HAS_MIRRORROW_NEON
ANY11M(MirrorRow_Any_NEON, MirrorRow_NEON, 1, 31)
#endif
#ifdef HAS_MIRRORROW_MSA
ANY11M(MirrorRow_Any_MSA, MirrorRow_MSA, 1, 63)
#endif
#ifdef HAS_MIRRORROW_MMI
ANY11M(MirrorRow_Any_MMI, MirrorRow_MMI, 1, 7)
#endif
#ifdef HAS_MIRRORUVROW_AVX2
ANY11M(MirrorUVRow_Any_AVX2, MirrorUVRow_AVX2, 2, 15)
#endif
#ifdef HAS_MIRRORUVROW_SSSE3
ANY11M(MirrorUVRow_Any_SSSE3, MirrorUVRow_SSSE3, 2, 7)
#endif
#ifdef HAS_MIRRORUVROW_NEON
ANY11M(MirrorUVRow_Any_NEON, MirrorUVRow_NEON, 2, 31)
#endif
#ifdef HAS_MIRRORUVROW_MSA
ANY11M(MirrorUVRow_Any_MSA, MirrorUVRow_MSA, 2, 7)
#endif
#ifdef HAS_ARGBMIRRORROW_AVX2
ANY11M(ARGBMirrorRow_Any_AVX2, ARGBMirrorRow_AVX2, 4, 7)
#endif
#ifdef HAS_ARGBMIRRORROW_SSE2
ANY11M(ARGBMirrorRow_Any_SSE2, ARGBMirrorRow_SSE2, 4, 3)
#endif
#ifdef HAS_ARGBMIRRORROW_NEON
ANY11M(ARGBMirrorRow_Any_NEON, ARGBMirrorRow_NEON, 4, 7)
#endif
#ifdef HAS_ARGBMIRRORROW_MSA
ANY11M(ARGBMirrorRow_Any_MSA, ARGBMirrorRow_MSA, 4, 15)
#endif
#ifdef HAS_ARGBMIRRORROW_MMI
ANY11M(ARGBMirrorRow_Any_MMI, ARGBMirrorRow_MMI, 4, 1)
#endif
#ifdef HAS_RGB24MIRRORROW_SSSE3
ANY11M(RGB24MirrorRow_Any_SSSE3, RGB24MirrorRow_SSSE3, 3, 15)
#endif
#ifdef HAS_RGB24MIRRORROW_NEON
ANY11M(RGB24MirrorRow_Any_NEON, RGB24MirrorRow_NEON, 3, 15)
#endif
#undef ANY11M
// Any 1 plane. (memset)
#define ANY1(NAMEANY, ANY_SIMD, T, BPP, MASK) \
void NAMEANY(uint8_t* dst_ptr, T v32, int width) { \
SIMD_ALIGNED(uint8_t temp[64]); \
memset(temp, 0, 64); /* for msan */ \
int r = width & MASK; \
int n = width & ~MASK; \
if (n > 0) { \
ANY_SIMD(dst_ptr, v32, n); \
} \
ANY_SIMD(temp, v32, MASK + 1); \
memcpy(dst_ptr + n * BPP, temp, r * BPP); \
}
#ifdef HAS_SETROW_X86
ANY1(SetRow_Any_X86, SetRow_X86, uint8_t, 1, 3)
#endif
#ifdef HAS_SETROW_NEON
ANY1(SetRow_Any_NEON, SetRow_NEON, uint8_t, 1, 15)
#endif
#ifdef HAS_ARGBSETROW_NEON
ANY1(ARGBSetRow_Any_NEON, ARGBSetRow_NEON, uint32_t, 4, 3)
#endif
#ifdef HAS_ARGBSETROW_MSA
ANY1(ARGBSetRow_Any_MSA, ARGBSetRow_MSA, uint32_t, 4, 3)
#endif
#ifdef HAS_ARGBSETROW_MMI
ANY1(ARGBSetRow_Any_MMI, ARGBSetRow_MMI, uint32_t, 4, 3)
#endif
#undef ANY1
// Any 1 to 2. Outputs UV planes.
#define ANY12(NAMEANY, ANY_SIMD, UVSHIFT, BPP, DUVSHIFT, MASK) \
void NAMEANY(const uint8_t* src_ptr, uint8_t* dst_u, uint8_t* dst_v, \
int width) { \
SIMD_ALIGNED(uint8_t temp[128 * 3]); \
memset(temp, 0, 128); /* for msan */ \
int r = width & MASK; \
int n = width & ~MASK; \
if (n > 0) { \
ANY_SIMD(src_ptr, dst_u, dst_v, n); \
} \
memcpy(temp, src_ptr + (n >> UVSHIFT) * BPP, SS(r, UVSHIFT) * BPP); \
ANY_SIMD(temp, temp + 128, temp + 256, MASK + 1); \
memcpy(dst_u + (n >> DUVSHIFT), temp + 128, SS(r, DUVSHIFT)); \
memcpy(dst_v + (n >> DUVSHIFT), temp + 256, SS(r, DUVSHIFT)); \
}
#ifdef HAS_SPLITUVROW_SSE2
ANY12(SplitUVRow_Any_SSE2, SplitUVRow_SSE2, 0, 2, 0, 15)
#endif
#ifdef HAS_SPLITUVROW_AVX2
ANY12(SplitUVRow_Any_AVX2, SplitUVRow_AVX2, 0, 2, 0, 31)
#endif
#ifdef HAS_SPLITUVROW_NEON
ANY12(SplitUVRow_Any_NEON, SplitUVRow_NEON, 0, 2, 0, 15)
#endif
#ifdef HAS_SPLITUVROW_MSA
ANY12(SplitUVRow_Any_MSA, SplitUVRow_MSA, 0, 2, 0, 31)
#endif
#ifdef HAS_SPLITUVROW_MMI
ANY12(SplitUVRow_Any_MMI, SplitUVRow_MMI, 0, 2, 0, 7)
#endif
#ifdef HAS_ARGBTOUV444ROW_SSSE3
ANY12(ARGBToUV444Row_Any_SSSE3, ARGBToUV444Row_SSSE3, 0, 4, 0, 15)
#endif
#ifdef HAS_YUY2TOUV422ROW_AVX2
ANY12(YUY2ToUV422Row_Any_AVX2, YUY2ToUV422Row_AVX2, 1, 4, 1, 31)
ANY12(UYVYToUV422Row_Any_AVX2, UYVYToUV422Row_AVX2, 1, 4, 1, 31)
#endif
#ifdef HAS_YUY2TOUV422ROW_SSE2
ANY12(YUY2ToUV422Row_Any_SSE2, YUY2ToUV422Row_SSE2, 1, 4, 1, 15)
ANY12(UYVYToUV422Row_Any_SSE2, UYVYToUV422Row_SSE2, 1, 4, 1, 15)
#endif
#ifdef HAS_YUY2TOUV422ROW_NEON
ANY12(ARGBToUV444Row_Any_NEON, ARGBToUV444Row_NEON, 0, 4, 0, 7)
ANY12(YUY2ToUV422Row_Any_NEON, YUY2ToUV422Row_NEON, 1, 4, 1, 15)
ANY12(UYVYToUV422Row_Any_NEON, UYVYToUV422Row_NEON, 1, 4, 1, 15)
#endif
#ifdef HAS_YUY2TOUV422ROW_MSA
ANY12(ARGBToUV444Row_Any_MSA, ARGBToUV444Row_MSA, 0, 4, 0, 15)
ANY12(YUY2ToUV422Row_Any_MSA, YUY2ToUV422Row_MSA, 1, 4, 1, 31)
ANY12(UYVYToUV422Row_Any_MSA, UYVYToUV422Row_MSA, 1, 4, 1, 31)
#endif
#ifdef HAS_YUY2TOUV422ROW_MMI
ANY12(ARGBToUV444Row_Any_MMI, ARGBToUV444Row_MMI, 0, 4, 0, 7)
ANY12(UYVYToUV422Row_Any_MMI, UYVYToUV422Row_MMI, 1, 4, 1, 15)
ANY12(YUY2ToUV422Row_Any_MMI, YUY2ToUV422Row_MMI, 1, 4, 1, 15)
#endif
#undef ANY12
// Any 1 to 3. Outputs RGB planes.
#define ANY13(NAMEANY, ANY_SIMD, BPP, MASK) \
void NAMEANY(const uint8_t* src_ptr, uint8_t* dst_r, uint8_t* dst_g, \
uint8_t* dst_b, int width) { \
SIMD_ALIGNED(uint8_t temp[16 * 6]); \
memset(temp, 0, 16 * 3); /* for msan */ \
int r = width & MASK; \
int n = width & ~MASK; \
if (n > 0) { \
ANY_SIMD(src_ptr, dst_r, dst_g, dst_b, n); \
} \
memcpy(temp, src_ptr + n * BPP, r * BPP); \
ANY_SIMD(temp, temp + 16 * 3, temp + 16 * 4, temp + 16 * 5, MASK + 1); \
memcpy(dst_r + n, temp + 16 * 3, r); \
memcpy(dst_g + n, temp + 16 * 4, r); \
memcpy(dst_b + n, temp + 16 * 5, r); \
}
#ifdef HAS_SPLITRGBROW_SSSE3
ANY13(SplitRGBRow_Any_SSSE3, SplitRGBRow_SSSE3, 3, 15)
#endif
#ifdef HAS_SPLITRGBROW_NEON
ANY13(SplitRGBRow_Any_NEON, SplitRGBRow_NEON, 3, 15)
#endif
#ifdef HAS_SPLITRGBROW_MMI
ANY13(SplitRGBRow_Any_MMI, SplitRGBRow_MMI, 3, 3)
#endif
// Any 1 to 2 with source stride (2 rows of source). Outputs UV planes.
// 128 byte row allows for 32 avx ARGB pixels.
#define ANY12S(NAMEANY, ANY_SIMD, UVSHIFT, BPP, MASK) \
void NAMEANY(const uint8_t* src_ptr, int src_stride_ptr, uint8_t* dst_u, \
uint8_t* dst_v, int width) { \
SIMD_ALIGNED(uint8_t temp[128 * 4]); \
memset(temp, 0, 128 * 2); /* for msan */ \
int r = width & MASK; \
int n = width & ~MASK; \
if (n > 0) { \
ANY_SIMD(src_ptr, src_stride_ptr, dst_u, dst_v, n); \
} \
memcpy(temp, src_ptr + (n >> UVSHIFT) * BPP, SS(r, UVSHIFT) * BPP); \
memcpy(temp + 128, src_ptr + src_stride_ptr + (n >> UVSHIFT) * BPP, \
SS(r, UVSHIFT) * BPP); \
if ((width & 1) && UVSHIFT == 0) { /* repeat last pixel for subsample */ \
memcpy(temp + SS(r, UVSHIFT) * BPP, temp + SS(r, UVSHIFT) * BPP - BPP, \
BPP); \
memcpy(temp + 128 + SS(r, UVSHIFT) * BPP, \
temp + 128 + SS(r, UVSHIFT) * BPP - BPP, BPP); \
} \
ANY_SIMD(temp, 128, temp + 256, temp + 384, MASK + 1); \
memcpy(dst_u + (n >> 1), temp + 256, SS(r, 1)); \
memcpy(dst_v + (n >> 1), temp + 384, SS(r, 1)); \
}
#ifdef HAS_ARGBTOUVROW_AVX2
ANY12S(ARGBToUVRow_Any_AVX2, ARGBToUVRow_AVX2, 0, 4, 31)
#endif
#ifdef HAS_ABGRTOUVROW_AVX2
ANY12S(ABGRToUVRow_Any_AVX2, ABGRToUVRow_AVX2, 0, 4, 31)
#endif
#ifdef HAS_ARGBTOUVJROW_AVX2
ANY12S(ARGBToUVJRow_Any_AVX2, ARGBToUVJRow_AVX2, 0, 4, 31)
#endif
#ifdef HAS_ARGBTOUVROW_SSSE3
ANY12S(ARGBToUVRow_Any_SSSE3, ARGBToUVRow_SSSE3, 0, 4, 15)
ANY12S(ARGBToUVJRow_Any_SSSE3, ARGBToUVJRow_SSSE3, 0, 4, 15)
ANY12S(BGRAToUVRow_Any_SSSE3, BGRAToUVRow_SSSE3, 0, 4, 15)
ANY12S(ABGRToUVRow_Any_SSSE3, ABGRToUVRow_SSSE3, 0, 4, 15)
ANY12S(RGBAToUVRow_Any_SSSE3, RGBAToUVRow_SSSE3, 0, 4, 15)
#endif
#ifdef HAS_YUY2TOUVROW_AVX2
ANY12S(YUY2ToUVRow_Any_AVX2, YUY2ToUVRow_AVX2, 1, 4, 31)
ANY12S(UYVYToUVRow_Any_AVX2, UYVYToUVRow_AVX2, 1, 4, 31)
#endif
#ifdef HAS_YUY2TOUVROW_SSE2
ANY12S(YUY2ToUVRow_Any_SSE2, YUY2ToUVRow_SSE2, 1, 4, 15)
ANY12S(UYVYToUVRow_Any_SSE2, UYVYToUVRow_SSE2, 1, 4, 15)
#endif
#ifdef HAS_ARGBTOUVROW_NEON
ANY12S(ARGBToUVRow_Any_NEON, ARGBToUVRow_NEON, 0, 4, 15)
#endif
#ifdef HAS_ARGBTOUVROW_MSA
ANY12S(ARGBToUVRow_Any_MSA, ARGBToUVRow_MSA, 0, 4, 31)
#endif
#ifdef HAS_ARGBTOUVROW_MMI
ANY12S(ARGBToUVRow_Any_MMI, ARGBToUVRow_MMI, 0, 4, 15)
#endif
#ifdef HAS_ARGBTOUVJROW_NEON
ANY12S(ARGBToUVJRow_Any_NEON, ARGBToUVJRow_NEON, 0, 4, 15)
#endif
#ifdef HAS_ARGBTOUVJROW_MSA
ANY12S(ARGBToUVJRow_Any_MSA, ARGBToUVJRow_MSA, 0, 4, 31)
#endif
#ifdef HAS_ARGBTOUVJROW_MMI
ANY12S(ARGBToUVJRow_Any_MMI, ARGBToUVJRow_MMI, 0, 4, 15)
#endif
#ifdef HAS_BGRATOUVROW_NEON
ANY12S(BGRAToUVRow_Any_NEON, BGRAToUVRow_NEON, 0, 4, 15)
#endif
#ifdef HAS_BGRATOUVROW_MSA
ANY12S(BGRAToUVRow_Any_MSA, BGRAToUVRow_MSA, 0, 4, 15)
#endif
#ifdef HAS_BGRATOUVROW_MMI
ANY12S(BGRAToUVRow_Any_MMI, BGRAToUVRow_MMI, 0, 4, 15)
#endif
#ifdef HAS_ABGRTOUVROW_NEON
ANY12S(ABGRToUVRow_Any_NEON, ABGRToUVRow_NEON, 0, 4, 15)
#endif
#ifdef HAS_ABGRTOUVROW_MSA
ANY12S(ABGRToUVRow_Any_MSA, ABGRToUVRow_MSA, 0, 4, 15)
#endif
#ifdef HAS_ABGRTOUVROW_MMI
ANY12S(ABGRToUVRow_Any_MMI, ABGRToUVRow_MMI, 0, 4, 15)
#endif
#ifdef HAS_RGBATOUVROW_NEON
ANY12S(RGBAToUVRow_Any_NEON, RGBAToUVRow_NEON, 0, 4, 15)
#endif
#ifdef HAS_RGBATOUVROW_MSA
ANY12S(RGBAToUVRow_Any_MSA, RGBAToUVRow_MSA, 0, 4, 15)
#endif
#ifdef HAS_RGBATOUVROW_MMI
ANY12S(RGBAToUVRow_Any_MMI, RGBAToUVRow_MMI, 0, 4, 15)
#endif
#ifdef HAS_RGB24TOUVROW_NEON
ANY12S(RGB24ToUVRow_Any_NEON, RGB24ToUVRow_NEON, 0, 3, 15)
#endif
#ifdef HAS_RGB24TOUVROW_MSA
ANY12S(RGB24ToUVRow_Any_MSA, RGB24ToUVRow_MSA, 0, 3, 15)
#endif
#ifdef HAS_RGB24TOUVROW_MMI
ANY12S(RGB24ToUVRow_Any_MMI, RGB24ToUVRow_MMI, 0, 3, 15)
#endif
#ifdef HAS_RAWTOUVROW_NEON
ANY12S(RAWToUVRow_Any_NEON, RAWToUVRow_NEON, 0, 3, 15)
#endif
#ifdef HAS_RAWTOUVROW_MSA
ANY12S(RAWToUVRow_Any_MSA, RAWToUVRow_MSA, 0, 3, 15)
#endif
#ifdef HAS_RAWTOUVROW_MMI
ANY12S(RAWToUVRow_Any_MMI, RAWToUVRow_MMI, 0, 3, 15)
#endif
#ifdef HAS_RGB565TOUVROW_NEON
ANY12S(RGB565ToUVRow_Any_NEON, RGB565ToUVRow_NEON, 0, 2, 15)
#endif
#ifdef HAS_RGB565TOUVROW_MSA
ANY12S(RGB565ToUVRow_Any_MSA, RGB565ToUVRow_MSA, 0, 2, 15)
#endif
#ifdef HAS_RGB565TOUVROW_MMI
ANY12S(RGB565ToUVRow_Any_MMI, RGB565ToUVRow_MMI, 0, 2, 15)
#endif
#ifdef HAS_ARGB1555TOUVROW_NEON
ANY12S(ARGB1555ToUVRow_Any_NEON, ARGB1555ToUVRow_NEON, 0, 2, 15)
#endif
#ifdef HAS_ARGB1555TOUVROW_MSA
ANY12S(ARGB1555ToUVRow_Any_MSA, ARGB1555ToUVRow_MSA, 0, 2, 15)
#endif
#ifdef HAS_ARGB1555TOUVROW_MMI
ANY12S(ARGB1555ToUVRow_Any_MMI, ARGB1555ToUVRow_MMI, 0, 2, 15)
#endif
#ifdef HAS_ARGB4444TOUVROW_NEON
ANY12S(ARGB4444ToUVRow_Any_NEON, ARGB4444ToUVRow_NEON, 0, 2, 15)
#endif
#ifdef HAS_ARGB4444TOUVROW_MMI
ANY12S(ARGB4444ToUVRow_Any_MMI, ARGB4444ToUVRow_MMI, 0, 2, 15)
#endif
#ifdef HAS_YUY2TOUVROW_NEON
ANY12S(YUY2ToUVRow_Any_NEON, YUY2ToUVRow_NEON, 1, 4, 15)
#endif
#ifdef HAS_UYVYTOUVROW_NEON
ANY12S(UYVYToUVRow_Any_NEON, UYVYToUVRow_NEON, 1, 4, 15)
#endif
#ifdef HAS_YUY2TOUVROW_MSA
ANY12S(YUY2ToUVRow_Any_MSA, YUY2ToUVRow_MSA, 1, 4, 31)
#endif
#ifdef HAS_YUY2TOUVROW_MMI
ANY12S(YUY2ToUVRow_Any_MMI, YUY2ToUVRow_MMI, 1, 4, 15)
#endif
#ifdef HAS_UYVYTOUVROW_MSA
ANY12S(UYVYToUVRow_Any_MSA, UYVYToUVRow_MSA, 1, 4, 31)
#endif
#ifdef HAS_UYVYTOUVROW_MMI
ANY12S(UYVYToUVRow_Any_MMI, UYVYToUVRow_MMI, 1, 4, 15)
#endif
#undef ANY12S
// Any 1 to 1 with source stride (2 rows of source). Outputs UV plane.
// 128 byte row allows for 32 avx ARGB pixels.
#define ANY11S(NAMEANY, ANY_SIMD, UVSHIFT, BPP, MASK) \
void NAMEANY(const uint8_t* src_ptr, int src_stride_ptr, uint8_t* dst_vu, \
int width) { \
SIMD_ALIGNED(uint8_t temp[128 * 3]); \
memset(temp, 0, 128 * 2); /* for msan */ \
int r = width & MASK; \
int n = width & ~MASK; \
if (n > 0) { \
ANY_SIMD(src_ptr, src_stride_ptr, dst_vu, n); \
} \
memcpy(temp, src_ptr + (n >> UVSHIFT) * BPP, SS(r, UVSHIFT) * BPP); \
memcpy(temp + 128, src_ptr + src_stride_ptr + (n >> UVSHIFT) * BPP, \
SS(r, UVSHIFT) * BPP); \
if ((width & 1) && UVSHIFT == 0) { /* repeat last pixel for subsample */ \
memcpy(temp + SS(r, UVSHIFT) * BPP, temp + SS(r, UVSHIFT) * BPP - BPP, \
BPP); \
memcpy(temp + 128 + SS(r, UVSHIFT) * BPP, \
temp + 128 + SS(r, UVSHIFT) * BPP - BPP, BPP); \
} \
ANY_SIMD(temp, 128, temp + 256, MASK + 1); \
memcpy(dst_vu + (n >> 1) * 2, temp + 256, SS(r, 1) * 2); \
}
#ifdef HAS_AYUVTOVUROW_NEON
ANY11S(AYUVToUVRow_Any_NEON, AYUVToUVRow_NEON, 0, 4, 15)
ANY11S(AYUVToVURow_Any_NEON, AYUVToVURow_NEON, 0, 4, 15)
#endif
#undef ANY11S
#ifdef __cplusplus
} // extern "C"
} // namespace libyuv
#endif
|
