aboutsummaryrefslogtreecommitdiffstats
path: root/libavcodec/x86/hevc_mc.asm
blob: b3b589b2713af482c9a2e8ebdac3f7d9c73ad1c0 (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
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
; /*
; * Provide SSE luma and chroma mc functions for HEVC decoding
; * Copyright (c) 2013 Pierre-Edouard LEPERE
; *
; * This file is part of FFmpeg.
; *
; * FFmpeg is free software; you can redistribute it and/or
; * modify it under the terms of the GNU Lesser General Public
; * License as published by the Free Software Foundation; either
; * version 2.1 of the License, or (at your option) any later version.
; *
; * FFmpeg is distributed in the hope that it will be useful,
; * but WITHOUT ANY WARRANTY; without even the implied warranty of
; * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
; * Lesser General Public License for more details.
; *
; * You should have received a copy of the GNU Lesser General Public
; * License along with FFmpeg; if not, write to the Free Software
; * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
; */
%include "libavutil/x86/x86util.asm"

SECTION_RODATA 32
cextern pw_255
cextern pw_512
cextern pw_2048
cextern pw_8192
cextern pw_1023
cextern pw_1024
cextern pw_4096
%define pw_8 pw_512
%define pw_10 pw_2048
%define pw_12 pw_8192
%define pw_bi_10 pw_1024
%define pw_bi_12 pw_4096
%define max_pixels_8 pw_255
%define max_pixels_10 pw_1023
pw_bi_8:                times 16 dw  (1 <<  8)
max_pixels_12:          times 16 dw ((1 << 12)-1)
cextern pd_1
cextern pb_0

%macro EPEL_TABLE 4
hevc_epel_filters_%4_%1 times %2 d%3 -2, 58
                        times %2 d%3 10, -2
                        times %2 d%3 -4, 54
                        times %2 d%3 16, -2
                        times %2 d%3 -6, 46
                        times %2 d%3 28, -4
                        times %2 d%3 -4, 36
                        times %2 d%3 36, -4
                        times %2 d%3 -4, 28
                        times %2 d%3 46, -6
                        times %2 d%3 -2, 16
                        times %2 d%3 54, -4
                        times %2 d%3 -2, 10
                        times %2 d%3 58, -2
%endmacro


EPEL_TABLE  8,16, b, avx2
EPEL_TABLE 10, 8, w, avx2

EPEL_TABLE  8, 8, b, sse4
EPEL_TABLE 10, 4, w, sse4
EPEL_TABLE 12, 4, w, sse4

%macro QPEL_TABLE 4
hevc_qpel_filters_%4_%1 times %2 d%3  -1,  4
                        times %2 d%3 -10, 58
                        times %2 d%3  17, -5
                        times %2 d%3   1,  0
                        times %2 d%3  -1,  4
                        times %2 d%3 -11, 40
                        times %2 d%3  40,-11
                        times %2 d%3   4, -1
                        times %2 d%3   0,  1
                        times %2 d%3  -5, 17
                        times %2 d%3  58,-10
                        times %2 d%3   4, -1
%endmacro

QPEL_TABLE  8, 8, b, sse4
QPEL_TABLE 10, 4, w, sse4
QPEL_TABLE 12, 4, w, sse4

QPEL_TABLE  8,16, b, avx2
QPEL_TABLE 10, 8, w, avx2

QPEL_TABLE  4, 1, b, avx512icl_h
QPEL_TABLE  8, 1, b, avx512icl_h
QPEL_TABLE  8, 1, d, avx512icl_v
QPEL_TABLE 16, 1, b, avx512icl_h
QPEL_TABLE 32, 1, b, avx512icl_h
QPEL_TABLE 64, 1, b, avx512icl_h

pb_qpel_shuffle_index: db  0,  1,  2,  3
                       db  1,  2,  3,  4
                       db  2,  3,  4,  5
                       db  3,  4,  5,  6
                       db  4,  5,  6,  7
                       db  5,  6,  7,  8
                       db  6,  7,  8,  9
                       db  7,  8,  9, 10
                       db  8,  9, 10, 11
                       db  9, 10, 11, 12
                       db 10, 11, 12, 13
                       db 11, 12, 13, 14
                       db 12, 13, 14, 15
                       db 13, 14, 15, 16
                       db 14, 15, 16, 17
                       db 15, 16, 17, 18
                       db  4,  5,  6,  7
                       db  5,  6,  7,  8
                       db  6,  7,  8,  9
                       db  7,  8,  9, 10
                       db  8,  9, 10, 11
                       db  9, 10, 11, 12
                       db 10, 11, 12, 13
                       db 11, 12, 13, 14
                       db 12, 13, 14, 15
                       db 13, 14, 15, 16
                       db 14, 15, 16, 17
                       db 15, 16, 17, 18
                       db 16, 17, 18, 19
                       db 17, 18, 19, 20
                       db 18, 19, 20, 21
                       db 19, 20, 21, 22

SECTION .text

%define MAX_PB_SIZE  64

%define hevc_qpel_filters_sse4_14 hevc_qpel_filters_sse4_10

%define hevc_qpel_filters_avx2_14 hevc_qpel_filters_avx2_10

%if ARCH_X86_64

%macro SIMPLE_BILOAD 4   ;width, tab, r1, r2
%if %1 <= 4
    movq              %3, [%2]                                              ; load data from source2
%elif %1 <= 8
    movdqa            %3, [%2]                                              ; load data from source2
%elif %1 <= 12
%if cpuflag(avx2)
    mova              %3, [%2]
%else
    movdqa            %3, [%2]                                              ; load data from source2
    movq              %4, [%2+16]                                           ; load data from source2
%endif ;avx
%elif %1 <= 16
%if cpuflag(avx2)
    mova              %3, [%2]
%else
    movdqa            %3, [%2]                                              ; load data from source2
    movdqa            %4, [%2+16]                                           ; load data from source2
%endif ; avx
%else ; %1 = 32
    mova              %3, [%2]
    mova              %4, [%2+32]
%endif
%endmacro

%macro SIMPLE_LOAD 4    ;width, bitd, tab, r1
%if %1 == 2 || (%2 == 8 && %1 <= 4)
    movd              %4, [%3]                                               ; load data from source
%elif %1 == 4 || (%2 == 8 && %1 <= 8)
    movq              %4, [%3]                                               ; load data from source
%elif notcpuflag(avx)
    movu              %4, [%3]                                               ; load data from source
%elif %1 <= 8 || (%2 == 8 && %1 <= 16)
    movdqu           %4, [%3]
%else
    movu              %4, [%3]
%endif
%endmacro


%macro EPEL_FILTER 5 ; bit depth, filter index, xmma, xmmb, gprtmp
%if cpuflag(avx2)
%assign %%offset 32
%if PIC
    lea              %5q, [hevc_epel_filters_avx2_%1]
    %define FILTER %5q
%else
    %define FILTER hevc_epel_filters_avx2_%1
%endif
%else
%assign %%offset 16
%if PIC
    lea              %5q, [hevc_epel_filters_sse4_%1]
    %define FILTER %5q
%else
    %define FILTER hevc_epel_filters_sse4_%1
%endif
%endif ;cpuflag(avx2)
    sub              %2q, 1
%if cpuflag(avx2)
    shl              %2q, 6                      ; multiply by 64
  %else
    shl              %2q, 5                      ; multiply by 32
%endif
    mova           %3, [FILTER + %2q]        ; get 2 first values of filters
    mova           %4, [FILTER + %2q+%%offset]     ; get 2 last values of filters
%endmacro

%macro EPEL_HV_FILTER 1
%if cpuflag(avx2)
%assign %%offset 32
%assign %%shift  6
%define %%table  hevc_epel_filters_avx2_%1
%else
%assign %%offset 16
%assign %%shift  5
%define %%table  hevc_epel_filters_sse4_%1
%endif

%if PIC
    lea           r3srcq, [%%table]
    %define FILTER r3srcq
%else
    %define FILTER %%table
%endif
    sub              mxq, 1
    sub              myq, 1
    shl              mxq, %%shift                ; multiply by 32
    shl              myq, %%shift                ; multiply by 32
    mova             m14, [FILTER + mxq]        ; get 2 first values of filters
    mova             m15, [FILTER + mxq+%%offset]     ; get 2 last values of filters

%if cpuflag(avx2)
%define %%table  hevc_epel_filters_avx2_10
%else
%define %%table  hevc_epel_filters_sse4_10
%endif
%if PIC
    lea           r3srcq, [%%table]
    %define FILTER r3srcq
%else
    %define FILTER %%table
%endif
    mova             m12, [FILTER + myq]        ; get 2 first values of filters
    mova             m13, [FILTER + myq+%%offset]     ; get 2 last values of filters
    lea           r3srcq, [srcstrideq*3]
%endmacro

%macro QPEL_FILTER 2

%if cpuflag(avx2)
%assign %%offset 32
%assign %%shift  7
%define %%table  hevc_qpel_filters_avx2_%1
%else
%assign %%offset 16
%assign %%shift  6
%define %%table  hevc_qpel_filters_sse4_%1
%endif

%if PIC
    lea         rfilterq, [%%table]
%else
    %define rfilterq %%table
%endif
    sub              %2q, 1
    shl              %2q, %%shift                        ; multiply by 32
    mova             m12, [rfilterq + %2q]               ; get 4 first values of filters
    mova             m13, [rfilterq + %2q +   %%offset]  ; get 4 first values of filters
    mova             m14, [rfilterq + %2q + 2*%%offset]  ; get 4 first values of filters
    mova             m15, [rfilterq + %2q + 3*%%offset]  ; get 4 first values of filters
%endmacro

%macro EPEL_LOAD 4
%if (%1 == 8 && %4 <= 4)
%define %%load movd
%elif (%1 == 8 && %4 <= 8) || (%1 > 8 && %4 <= 4)
%define %%load movq
%else
%define %%load movdqu
%endif

    %%load            m0, [%2q ]
%ifnum %3
    %%load            m1, [%2q+  %3]
    %%load            m2, [%2q+2*%3]
    %%load            m3, [%2q+3*%3]
%else
    %%load            m1, [%2q+  %3q]
    %%load            m2, [%2q+2*%3q]
    %%load            m3, [%2q+r3srcq]
%endif
%if %1 == 8
%if %4 > 8
    SBUTTERFLY        bw, 0, 1, 7
    SBUTTERFLY        bw, 2, 3, 7
%else
    punpcklbw         m0, m1
    punpcklbw         m2, m3
%endif
%else
%if %4 > 4
    SBUTTERFLY        wd, 0, 1, 7
    SBUTTERFLY        wd, 2, 3, 7
%else
    punpcklwd         m0, m1
    punpcklwd         m2, m3
%endif
%endif
%endmacro


%macro QPEL_H_LOAD 4
%assign %%stride (%1+7)/8
%if %1 == 8
%if %3 <= 4
%define %%load movd
%elif %3 == 8
%define %%load movq
%else
%define %%load movu
%endif
%else
%if %3 == 2
%define %%load movd
%elif %3 == 4
%define %%load movq
%else
%define %%load movu
%endif
%endif
    %%load            m0, [%2-3*%%stride]        ;load data from source
    %%load            m1, [%2-2*%%stride]
    %%load            m2, [%2-%%stride  ]
    %%load            m3, [%2           ]
    %%load            m4, [%2+%%stride  ]
    %%load            m5, [%2+2*%%stride]
    %%load            m6, [%2+3*%%stride]
    %%load            m7, [%2+4*%%stride]

%if %1 == 8
%if %3 > 8
    SBUTTERFLY        wd, 0, 1, %4
    SBUTTERFLY        wd, 2, 3, %4
    SBUTTERFLY        wd, 4, 5, %4
    SBUTTERFLY        wd, 6, 7, %4
%else
    punpcklbw         m0, m1
    punpcklbw         m2, m3
    punpcklbw         m4, m5
    punpcklbw         m6, m7
%endif
%else
%if %3 > 4
    SBUTTERFLY        dq, 0, 1, %4
    SBUTTERFLY        dq, 2, 3, %4
    SBUTTERFLY        dq, 4, 5, %4
    SBUTTERFLY        dq, 6, 7, %4
%else
    punpcklwd         m0, m1
    punpcklwd         m2, m3
    punpcklwd         m4, m5
    punpcklwd         m6, m7
%endif
%endif
%endmacro

%macro QPEL_V_LOAD 5
    lea              %5q, [%2]
    sub              %5q, r3srcq
    movu              m0, [%5q            ]      ;load x- 3*srcstride
    movu              m1, [%5q+   %3q     ]      ;load x- 2*srcstride
    movu              m2, [%5q+ 2*%3q     ]      ;load x-srcstride
    movu              m3, [%2       ]      ;load x
    movu              m4, [%2+   %3q]      ;load x+stride
    movu              m5, [%2+ 2*%3q]      ;load x+2*stride
    movu              m6, [%2+r3srcq]      ;load x+3*stride
    movu              m7, [%2+ 4*%3q]      ;load x+4*stride
%if %1 == 8
%if %4 > 8
    SBUTTERFLY        bw, 0, 1, 8
    SBUTTERFLY        bw, 2, 3, 8
    SBUTTERFLY        bw, 4, 5, 8
    SBUTTERFLY        bw, 6, 7, 8
%else
    punpcklbw         m0, m1
    punpcklbw         m2, m3
    punpcklbw         m4, m5
    punpcklbw         m6, m7
%endif
%else
%if %4 > 4
    SBUTTERFLY        wd, 0, 1, 8
    SBUTTERFLY        wd, 2, 3, 8
    SBUTTERFLY        wd, 4, 5, 8
    SBUTTERFLY        wd, 6, 7, 8
%else
    punpcklwd         m0, m1
    punpcklwd         m2, m3
    punpcklwd         m4, m5
    punpcklwd         m6, m7
%endif
%endif
%endmacro

%macro PEL_12STORE2 3
    movd           [%1], %2
%endmacro
%macro PEL_12STORE4 3
    movq           [%1], %2
%endmacro
%macro PEL_12STORE6 3
    movq           [%1], %2
    psrldq            %2, 8
    movd         [%1+8], %2
%endmacro
%macro PEL_12STORE8 3
    movdqa         [%1], %2
%endmacro
%macro PEL_12STORE12 3
    movdqa         [%1], %2
    movq        [%1+16], %3
%endmacro
%macro PEL_12STORE16 3
    PEL_12STORE8      %1, %2, %3
    movdqa       [%1+16], %3
%endmacro

%macro PEL_10STORE2 3
    movd           [%1], %2
%endmacro
%macro PEL_10STORE4 3
    movq           [%1], %2
%endmacro
%macro PEL_10STORE6 3
    movq           [%1], %2
    psrldq            %2, 8
    movd         [%1+8], %2
%endmacro
%macro PEL_10STORE8 3
    movdqa         [%1], %2
%endmacro
%macro PEL_10STORE12 3
    movdqa         [%1], %2
    movq        [%1+16], %3
%endmacro
%macro PEL_10STORE16 3
%if cpuflag(avx2)
    movu            [%1], %2
%else
    PEL_10STORE8      %1, %2, %3
    movdqa       [%1+16], %3
%endif
%endmacro

%macro PEL_10STORE32 3
    PEL_10STORE16     %1, %2, %3
    movu         [%1+32], %3
%endmacro

%macro PEL_8STORE2 3
    pextrw          [%1], %2, 0
%endmacro
%macro PEL_8STORE4 3
    movd            [%1], %2
%endmacro
%macro PEL_8STORE6 3
    movd            [%1], %2
    pextrw        [%1+4], %2, 2
%endmacro
%macro PEL_8STORE8 3
    movq           [%1], %2
%endmacro
%macro PEL_8STORE12 3
    movq            [%1], %2
    psrldq            %2, 8
    movd          [%1+8], %2
%endmacro
%macro PEL_8STORE16 3
%if cpuflag(avx2)
    movdqu        [%1], %2
%else
    mova          [%1], %2
%endif ; avx
%endmacro
%macro PEL_8STORE32 3
    movu          [%1], %2
%endmacro

%macro LOOP_END 3
    add              %1q, 2*MAX_PB_SIZE          ; dst += dststride
    add              %2q, %3q                    ; src += srcstride
    dec          heightd                         ; cmp height
    jnz               .loop                      ; height loop
%endmacro


%macro MC_PIXEL_COMPUTE 2-3 ;width, bitdepth
%if %2 == 8
%if cpuflag(avx2) && %0 ==3
%if %1 > 16
    vextracti128 xm1, m0, 1
    pmovzxbw      m1, xm1
    psllw         m1, 14-%2
%endif
    pmovzxbw      m0, xm0
%else ; not avx
%if %1 > 8
    punpckhbw     m1, m0, m2
    psllw         m1, 14-%2
%endif
    punpcklbw     m0, m2
%endif
%endif ;avx
    psllw         m0, 14-%2
%endmacro

%macro EPEL_COMPUTE 4-8 ; bitdepth, width, filter1, filter2, HV/m0, m2, m1, m3
%if %0 == 8
%define %%reg0 %5
%define %%reg2 %6
%define %%reg1 %7
%define %%reg3 %8
%else
%define %%reg0 m0
%define %%reg2 m2
%define %%reg1 m1
%define %%reg3 m3
%endif
%if %1 == 8
%if cpuflag(avx2) && (%0 == 5)
%if %2 > 16
    vperm2i128    m10, m0, m1, q0301
%endif
    vinserti128    m0, m0, xm1, 1
    mova           m1, m10
%if %2 > 16
    vperm2i128    m10, m2, m3, q0301
%endif
    vinserti128    m2, m2, xm3, 1
    mova           m3, m10
%endif
    pmaddubsw      %%reg0, %3   ;x1*c1+x2*c2
    pmaddubsw      %%reg2, %4   ;x3*c3+x4*c4
    paddw          %%reg0, %%reg2
%if %2 > 8
    pmaddubsw      %%reg1, %3
    pmaddubsw      %%reg3, %4
    paddw          %%reg1, %%reg3
%endif
%else
    pmaddwd        %%reg0, %3
    pmaddwd        %%reg2, %4
    paddd          %%reg0, %%reg2
%if %2 > 4
    pmaddwd        %%reg1, %3
    pmaddwd        %%reg3, %4
    paddd          %%reg1, %%reg3
%if %1 != 8
    psrad          %%reg1, %1-8
%endif
%endif
%if %1 != 8
    psrad          %%reg0, %1-8
%endif
    packssdw       %%reg0, %%reg1
%endif
%endmacro

%macro QPEL_HV_COMPUTE 4     ; width, bitdepth, filter idx

%if cpuflag(avx2)
%assign %%offset 32
%define %%table  hevc_qpel_filters_avx2_%2
%else
%assign %%offset 16
%define %%table  hevc_qpel_filters_sse4_%2
%endif

%if PIC
    lea         rfilterq, [%%table]
%else
    %define rfilterq %%table
%endif

%if %2 == 8
    pmaddubsw         m0, [rfilterq + %3q*8   ]   ;x1*c1+x2*c2
    pmaddubsw         m2, [rfilterq + %3q*8+%%offset]   ;x3*c3+x4*c4
    pmaddubsw         m4, [rfilterq + %3q*8+2*%%offset]   ;x5*c5+x6*c6
    pmaddubsw         m6, [rfilterq + %3q*8+3*%%offset]   ;x7*c7+x8*c8
    paddw             m0, m2
    paddw             m4, m6
    paddw             m0, m4
%else
    pmaddwd           m0, [rfilterq + %3q*8   ]
    pmaddwd           m2, [rfilterq + %3q*8+%%offset]
    pmaddwd           m4, [rfilterq + %3q*8+2*%%offset]
    pmaddwd           m6, [rfilterq + %3q*8+3*%%offset]
    paddd             m0, m2
    paddd             m4, m6
    paddd             m0, m4
%if %2 != 8
    psrad             m0, %2-8
%endif
%if %1 > 4
    pmaddwd           m1, [rfilterq + %3q*8   ]
    pmaddwd           m3, [rfilterq + %3q*8+%%offset]
    pmaddwd           m5, [rfilterq + %3q*8+2*%%offset]
    pmaddwd           m7, [rfilterq + %3q*8+3*%%offset]
    paddd             m1, m3
    paddd             m5, m7
    paddd             m1, m5
%if %2 != 8
    psrad             m1, %2-8
%endif
%endif
    p%4               m0, m1
%endif
%endmacro

%macro QPEL_COMPUTE 2-3     ; width, bitdepth
%if %2 == 8
%if cpuflag(avx2) && (%0 == 3)

    vperm2i128 m10, m0,  m1, q0301
    vinserti128 m0, m0, xm1, 1
    SWAP 1, 10

    vperm2i128 m10, m2,  m3, q0301
    vinserti128 m2, m2, xm3, 1
    SWAP 3, 10


    vperm2i128 m10, m4,  m5, q0301
    vinserti128 m4, m4, xm5, 1
    SWAP 5, 10

    vperm2i128 m10, m6,  m7, q0301
    vinserti128 m6, m6, xm7, 1
    SWAP 7, 10
%endif

    pmaddubsw         m0, m12   ;x1*c1+x2*c2
    pmaddubsw         m2, m13   ;x3*c3+x4*c4
    pmaddubsw         m4, m14   ;x5*c5+x6*c6
    pmaddubsw         m6, m15   ;x7*c7+x8*c8
    paddw             m0, m2
    paddw             m4, m6
    paddw             m0, m4
%if %1 > 8
    pmaddubsw         m1, m12
    pmaddubsw         m3, m13
    pmaddubsw         m5, m14
    pmaddubsw         m7, m15
    paddw             m1, m3
    paddw             m5, m7
    paddw             m1, m5
%endif
%else
    pmaddwd           m0, m12
    pmaddwd           m2, m13
    pmaddwd           m4, m14
    pmaddwd           m6, m15
    paddd             m0, m2
    paddd             m4, m6
    paddd             m0, m4
%if %2 != 8
    psrad             m0, %2-8
%endif
%if %1 > 4
    pmaddwd           m1, m12
    pmaddwd           m3, m13
    pmaddwd           m5, m14
    pmaddwd           m7, m15
    paddd             m1, m3
    paddd             m5, m7
    paddd             m1, m5
%if %2 != 8
    psrad             m1, %2-8
%endif
%endif
%endif
%endmacro

%macro BI_COMPUTE 7-8     ; width, bitd, src1l, src1h, scr2l, scr2h, pw
    paddsw            %3, %5
%if %1 > 8
    paddsw            %4, %6
%endif
    UNI_COMPUTE       %1, %2, %3, %4, %7
%if %0 == 8 && cpuflag(avx2) && (%2 == 8)
    vpermq            %3, %3, 216
    vpermq            %4, %4, 216
%endif
%endmacro

%macro UNI_COMPUTE 5
    pmulhrsw          %3, %5
%if %1 > 8 || (%2 > 8 && %1 > 4)
    pmulhrsw          %4, %5
%endif
%if %2 == 8
    packuswb          %3, %4
%else
    CLIPW             %3, [pb_0], [max_pixels_%2]
%if (%1 > 8 && notcpuflag(avx)) || %1 > 16
    CLIPW             %4, [pb_0], [max_pixels_%2]
%endif
%endif
%endmacro


; ******************************
; void put_hevc_mc_pixels(int16_t *dst, ptrdiff_t dststride,
;                         const uint8_t *_src, ptrdiff_t _srcstride,
;                         int height, int mx, int my)
; ******************************

%macro HEVC_BI_PEL_PIXELS 2
cglobal hevc_put_hevc_bi_pel_pixels%1_%2, 6, 6, 6, dst, dststride, src, srcstride, src2, height
    pxor              m2, m2
    movdqa            m5, [pw_bi_%2]
.loop:
    SIMPLE_LOAD       %1, %2, srcq, m0
    SIMPLE_BILOAD     %1, src2q, m3, m4
    MC_PIXEL_COMPUTE  %1, %2, 1
    BI_COMPUTE        %1, %2, m0, m1, m3, m4, m5, 1
    PEL_%2STORE%1   dstq, m0, m1
    add             dstq, dststrideq             ; dst += dststride
    add             srcq, srcstrideq             ; src += srcstride
    add            src2q, 2*MAX_PB_SIZE          ; src += srcstride
    dec          heightd                         ; cmp height
    jnz               .loop                      ; height loop
    RET
%endmacro


; ******************************
; void put_hevc_epel_hX(int16_t *dst, ptrdiff_t dststride,
;                       const uint8_t *_src, ptrdiff_t _srcstride,
;                       int height, int mx, int my, int width);
; ******************************


%macro HEVC_PUT_HEVC_EPEL 2
%if cpuflag(avx2)
%define XMM_REGS  11
%else
%define XMM_REGS  8
%endif

cglobal hevc_put_hevc_bi_epel_h%1_%2, 7, 8, XMM_REGS, dst, dststride, src, srcstride, src2, height, mx, rfilter
%assign %%stride ((%2 + 7)/8)
    movdqa            m6, [pw_bi_%2]
    EPEL_FILTER       %2, mx, m4, m5, rfilter
.loop:
    EPEL_LOAD         %2, srcq-%%stride, %%stride, %1
    EPEL_COMPUTE      %2, %1, m4, m5, 1
    SIMPLE_BILOAD     %1, src2q, m2, m3
    BI_COMPUTE        %1, %2, m0, m1, m2, m3, m6, 1
    PEL_%2STORE%1   dstq, m0, m1
    add             dstq, dststrideq             ; dst += dststride
    add             srcq, srcstrideq             ; src += srcstride
    add            src2q, 2*MAX_PB_SIZE          ; src += srcstride
    dec          heightd                         ; cmp height
    jnz               .loop                      ; height loop
    RET

; ******************************
; void put_hevc_epel_v(int16_t *dst, ptrdiff_t dststride,
;                      const uint8_t *_src, ptrdiff_t _srcstride,
;                      int height, int mx, int my, int width)
; ******************************

cglobal hevc_put_hevc_bi_epel_v%1_%2, 6, 8, XMM_REGS, dst, dststride, src, srcstride, src2, height, r3src, my
    movifnidn        myd, mym
    movdqa            m6, [pw_bi_%2]
    sub             srcq, srcstrideq
    EPEL_FILTER       %2, my, m4, m5, r3src
    lea           r3srcq, [srcstrideq*3]
.loop:
    EPEL_LOAD         %2, srcq, srcstride, %1
    EPEL_COMPUTE      %2, %1, m4, m5, 1
    SIMPLE_BILOAD     %1, src2q, m2, m3
    BI_COMPUTE        %1, %2, m0, m1, m2, m3, m6, 1
    PEL_%2STORE%1   dstq, m0, m1
    add             dstq, dststrideq             ; dst += dststride
    add             srcq, srcstrideq             ; src += srcstride
    add            src2q, 2*MAX_PB_SIZE          ; src += srcstride
    dec          heightd                         ; cmp height
    jnz               .loop                      ; height loop
    RET
%endmacro


; ******************************
; void put_hevc_epel_hv(int16_t *dst, ptrdiff_t dststride,
;                       const uint8_t *_src, ptrdiff_t _srcstride,
;                       int height, int mx, int my, int width)
; ******************************

%macro HEVC_PUT_HEVC_EPEL_HV 2

cglobal hevc_put_hevc_bi_epel_hv%1_%2, 8, 9, 16, dst, dststride, src, srcstride, src2, height, mx, my, r3src
%assign %%stride ((%2 + 7)/8)
    sub             srcq, srcstrideq
    EPEL_HV_FILTER    %2
    EPEL_LOAD         %2, srcq-%%stride, %%stride, %1
    EPEL_COMPUTE      %2, %1, m14, m15
%if (%1 > 8 && (%2 == 8))
    SWAP              m8, m1
%endif
    SWAP              m4, m0
    add             srcq, srcstrideq
    EPEL_LOAD         %2, srcq-%%stride, %%stride, %1
    EPEL_COMPUTE      %2, %1, m14, m15
%if (%1 > 8 && (%2 == 8))
    SWAP              m9, m1
%endif
    SWAP              m5, m0
    add             srcq, srcstrideq
    EPEL_LOAD         %2, srcq-%%stride, %%stride, %1
    EPEL_COMPUTE      %2, %1, m14, m15
%if (%1 > 8 && (%2 == 8))
    SWAP             m10, m1
%endif
    SWAP              m6, m0
    add             srcq, srcstrideq
.loop:
    EPEL_LOAD         %2, srcq-%%stride, %%stride, %1
    EPEL_COMPUTE      %2, %1, m14, m15
%if (%1 > 8 && (%2 == 8))
    SWAP             m11, m1
%endif
    SWAP              m7, m0
    punpcklwd         m0, m4, m5
    punpcklwd         m2, m6, m7
%if %1 > 4
    punpckhwd         m1, m4, m5
    punpckhwd         m3, m6, m7
%endif
    EPEL_COMPUTE      14, %1, m12, m13
%if (%1 > 8 && (%2 == 8))
    punpcklwd         m4, m8, m9
    punpcklwd         m2, m10, m11
    punpckhwd         m8, m8, m9
    punpckhwd         m3, m10, m11
    EPEL_COMPUTE      14, %1, m12, m13, m4, m2, m8, m3
    SIMPLE_BILOAD     %1, src2q, m8, m3
%if cpuflag(avx2)
    vinserti128       m1, m8, xm3, 1
    vperm2i128        m2, m8, m3, q0301
    BI_COMPUTE        %1, %2, m0, m4, m1, m2, [pw_bi_%2]
%else
    BI_COMPUTE        %1, %2, m0, m4, m8, m3, [pw_bi_%2]
%endif
%else
    SIMPLE_BILOAD     %1, src2q, m8, m9
    BI_COMPUTE        %1, %2, m0, m1, m8, m9, [pw_bi_%2]
%endif
    PEL_%2STORE%1   dstq, m0, m4
    mova              m4, m5
    mova              m5, m6
    mova              m6, m7
%if (%1 > 8 && (%2 == 8))
    mova              m8, m9
    mova              m9, m10
    mova             m10, m11
%endif
    add             dstq, dststrideq             ; dst += dststride
    add             srcq, srcstrideq             ; src += srcstride
    add            src2q, 2*MAX_PB_SIZE          ; src += srcstride
    dec          heightd                         ; cmp height
    jnz               .loop                      ; height loop
    RET
%endmacro

; ******************************
; void put_hevc_qpel_hX_X_X(int16_t *dst, ptrdiff_t dststride,
;                           const uint8_t *_src, ptrdiff_t _srcstride,
;                       int height, int mx, int my, int width)
; ******************************

%macro HEVC_PUT_HEVC_QPEL 2

cglobal hevc_put_hevc_bi_qpel_h%1_%2, 7, 8, 16 , dst, dststride, src, srcstride, src2, height, mx, rfilter
    movdqa            m9, [pw_bi_%2]
    QPEL_FILTER       %2, mx
.loop:
    QPEL_H_LOAD       %2, srcq, %1, 10
    QPEL_COMPUTE      %1, %2, 1
%if %2 > 8
    packssdw          m0, m1
%endif
    SIMPLE_BILOAD     %1, src2q, m10, m11
    BI_COMPUTE        %1, %2, m0, m1, m10, m11, m9, 1
    PEL_%2STORE%1   dstq, m0, m1
    add             dstq, dststrideq             ; dst += dststride
    add             srcq, srcstrideq             ; src += srcstride
    add            src2q, 2*MAX_PB_SIZE          ; src += srcstride
    dec          heightd                         ; cmp height
    jnz               .loop                      ; height loop
    RET


; ******************************
; void put_hevc_qpel_vX_X_X(int16_t *dst, ptrdiff_t dststride,
;                           const uint8_t *_src, ptrdiff_t _srcstride,
;                       int height, int mx, int my, int width)
; ******************************


cglobal hevc_put_hevc_bi_qpel_v%1_%2, 6, 10, 16, dst, dststride, src, srcstride, src2, height, r3src, my, rfilter
    movifnidn        myd, mym
    movdqa            m9, [pw_bi_%2]
    lea           r3srcq, [srcstrideq*3]
    QPEL_FILTER       %2, my
.loop:
    QPEL_V_LOAD       %2, srcq, srcstride, %1, r9
    QPEL_COMPUTE      %1, %2, 1
%if %2 > 8
    packssdw          m0, m1
%endif
    SIMPLE_BILOAD     %1, src2q, m10, m11
    BI_COMPUTE        %1, %2, m0, m1, m10, m11, m9, 1
    PEL_%2STORE%1   dstq, m0, m1
    add             dstq, dststrideq             ; dst += dststride
    add             srcq, srcstrideq             ; src += srcstride
    add            src2q, 2*MAX_PB_SIZE          ; src += srcstride
    dec          heightd                         ; cmp height
    jnz               .loop                      ; height loop
    RET
%endmacro


; ******************************
; void put_hevc_qpel_hvX_X(int16_t *dst, ptrdiff_t dststride,
;                          const uint8_t *_src, ptrdiff_t _srcstride,
;                       int height, int mx, int my)
; ******************************
%macro HEVC_PUT_HEVC_QPEL_HV 2

cglobal hevc_put_hevc_bi_qpel_hv%1_%2, 8, 10, 16, dst, dststride, src, srcstride, src2, height, mx, my, r3src, rfilter
%if cpuflag(avx2)
%assign %%shift  4
%else
%assign %%shift  3
%endif
    sub              mxq, 1
    sub              myq, 1
    shl              mxq, %%shift                ; multiply by 32
    shl              myq, %%shift                ; multiply by 32
    lea           r3srcq, [srcstrideq*3]
    sub             srcq, r3srcq
    QPEL_H_LOAD       %2, srcq, %1, 15
    QPEL_HV_COMPUTE   %1, %2, mx, ackssdw
    SWAP              m8, m0
    add             srcq, srcstrideq
    QPEL_H_LOAD       %2, srcq, %1, 15
    QPEL_HV_COMPUTE   %1, %2, mx, ackssdw
    SWAP              m9, m0
    add             srcq, srcstrideq
    QPEL_H_LOAD       %2, srcq, %1, 15
    QPEL_HV_COMPUTE   %1, %2, mx, ackssdw
    SWAP             m10, m0
    add             srcq, srcstrideq
    QPEL_H_LOAD       %2, srcq, %1, 15
    QPEL_HV_COMPUTE   %1, %2, mx, ackssdw
    SWAP             m11, m0
    add             srcq, srcstrideq
    QPEL_H_LOAD       %2, srcq, %1, 15
    QPEL_HV_COMPUTE   %1, %2, mx, ackssdw
    SWAP             m12, m0
    add             srcq, srcstrideq
    QPEL_H_LOAD       %2, srcq, %1, 15
    QPEL_HV_COMPUTE   %1, %2, mx, ackssdw
    SWAP             m13, m0
    add             srcq, srcstrideq
    QPEL_H_LOAD       %2, srcq, %1, 15
    QPEL_HV_COMPUTE   %1, %2, mx, ackssdw
    SWAP             m14, m0
    add             srcq, srcstrideq
.loop:
    QPEL_H_LOAD       %2, srcq, %1, 15
    QPEL_HV_COMPUTE   %1, %2, mx, ackssdw
    SWAP             m15, m0
    punpcklwd         m0, m8, m9
    punpcklwd         m2, m10, m11
    punpcklwd         m4, m12, m13
    punpcklwd         m6, m14, m15
%if %1 > 4
    punpckhwd         m1, m8, m9
    punpckhwd         m3, m10, m11
    punpckhwd         m5, m12, m13
    punpckhwd         m7, m14, m15
%endif
    QPEL_HV_COMPUTE   %1, 14, my, ackssdw
    SIMPLE_BILOAD     %1, src2q, m8, m9 ;m9 not used in this case
    BI_COMPUTE        %1, %2, m0, m1, m8, m9, [pw_bi_%2]
    PEL_%2STORE%1   dstq, m0, m1

%if %1 <= 4
    movq              m8, m9
    movq              m9, m10
    movq             m10, m11
    movq             m11, m12
    movq             m12, m13
    movq             m13, m14
    movq             m14, m15
%else
    movdqa            m8, m9
    movdqa            m9, m10
    movdqa           m10, m11
    movdqa           m11, m12
    movdqa           m12, m13
    movdqa           m13, m14
    movdqa           m14, m15
%endif
    add             dstq, dststrideq             ; dst += dststride
    add             srcq, srcstrideq             ; src += srcstride
    add            src2q, 2*MAX_PB_SIZE          ; src += srcstride
    dec          heightd                         ; cmp height
    jnz               .loop                      ; height loop
    RET
%endmacro

%macro WEIGHTING_FUNCS 2
%if WIN64 || ARCH_X86_32
cglobal hevc_put_hevc_uni_w%1_%2, 4, 5, 7, dst, dststride, src, height, denom, wx, ox
    mov             r4d, denomm
%define SHIFT  r4d
%else
cglobal hevc_put_hevc_uni_w%1_%2, 6, 6, 7, dst, dststride, src, height, denom, wx, ox
%define SHIFT  denomd
%endif
    lea           SHIFT, [SHIFT+14-%2]          ; shift = 14 - bitd + denom
%if %1 <= 4
    pxor             m1, m1
%endif
    movd             m2, wxm        ; WX
    movd             m4, SHIFT      ; shift
%if %1 <= 4
    punpcklwd        m2, m1
%else
    punpcklwd        m2, m2
%endif
    dec           SHIFT
    movdqu           m5, [pd_1]
    movd             m6, SHIFT
    pshufd           m2, m2, 0
    mov           SHIFT, oxm
    pslld            m5, m6
%if %2 != 8
    shl           SHIFT, %2-8       ; ox << (bitd - 8)
%endif
    movd             m3, SHIFT      ; OX
    pshufd           m3, m3, 0
%if WIN64 || ARCH_X86_32
    mov           SHIFT, heightm
%endif
.loop:
   SIMPLE_LOAD        %1, 10, srcq, m0
%if %1 <= 4
    punpcklwd         m0, m1
    pmaddwd           m0, m2
    paddd             m0, m5
    psrad             m0, m4
    paddd             m0, m3
%else
    pmulhw            m6, m0, m2
    pmullw            m0, m2
    punpckhwd         m1, m0, m6
    punpcklwd         m0, m6
    paddd             m0, m5
    paddd             m1, m5
    psrad             m0, m4
    psrad             m1, m4
    paddd             m0, m3
    paddd             m1, m3
%endif
    packssdw          m0, m1
%if %2 == 8
    packuswb          m0, m0
%else
    CLIPW             m0, [pb_0], [max_pixels_%2]
%endif
    PEL_%2STORE%1   dstq, m0, m1
    add             dstq, dststrideq             ; dst += dststride
    add             srcq, 2*MAX_PB_SIZE          ; src += srcstride
    dec          heightd                         ; cmp height
    jnz               .loop                      ; height loop
    RET

cglobal hevc_put_hevc_bi_w%1_%2, 4, 6, 10, dst, dststride, src, src2, height, denom, wx0, wx1, ox0, ox1
    movifnidn        r5d, denomm
%if %1 <= 4
    pxor              m1, m1
%endif
    movd              m2, wx0m         ; WX0
    lea              r5d, [r5d+14-%2]  ; shift = 14 - bitd + denom
    movd              m3, wx1m         ; WX1
    movd              m0, r5d          ; shift
%if %1 <= 4
    punpcklwd         m2, m1
    punpcklwd         m3, m1
%else
    punpcklwd         m2, m2
    punpcklwd         m3, m3
%endif
    inc              r5d
    movd              m5, r5d          ; shift+1
    pshufd            m2, m2, 0
    mov              r5d, ox0m
    pshufd            m3, m3, 0
    add              r5d, ox1m
%if %2 != 8
    shl              r5d, %2-8         ; ox << (bitd - 8)
%endif
    inc              r5d
    movd              m4, r5d          ; offset
    pshufd            m4, m4, 0
%if UNIX64
%define h heightd
%else
    mov              r5d, heightm
%define h r5d
%endif
    pslld             m4, m0

.loop:
   SIMPLE_LOAD        %1, 10, srcq,  m0
   SIMPLE_LOAD        %1, 10, src2q, m8
%if %1 <= 4
    punpcklwd         m0, m1
    punpcklwd         m8, m1
    pmaddwd           m0, m3
    pmaddwd           m8, m2
    paddd             m0, m4
    paddd             m0, m8
    psrad             m0, m5
%else
    pmulhw            m6, m0, m3
    pmullw            m0, m3
    pmulhw            m7, m8, m2
    pmullw            m8, m2
    punpckhwd         m1, m0, m6
    punpcklwd         m0, m6
    punpckhwd         m9, m8, m7
    punpcklwd         m8, m7
    paddd             m0, m8
    paddd             m1, m9
    paddd             m0, m4
    paddd             m1, m4
    psrad             m0, m5
    psrad             m1, m5
%endif
    packssdw          m0, m1
%if %2 == 8
    packuswb          m0, m0
%else
     CLIPW            m0, [pb_0], [max_pixels_%2]
%endif
    PEL_%2STORE%1   dstq, m0, m1
    add             dstq, dststrideq             ; dst += dststride
    add             srcq, 2*MAX_PB_SIZE          ; src += srcstride
    add            src2q, 2*MAX_PB_SIZE          ; src2 += srcstride
    dec                h                         ; cmp height
    jnz               .loop                      ; height loop
    RET
%endmacro

INIT_XMM sse4                                    ; adds ff_ and _sse4 to function name

WEIGHTING_FUNCS 2, 8
WEIGHTING_FUNCS 4, 8
WEIGHTING_FUNCS 6, 8
WEIGHTING_FUNCS 8, 8

WEIGHTING_FUNCS 2, 10
WEIGHTING_FUNCS 4, 10
WEIGHTING_FUNCS 6, 10
WEIGHTING_FUNCS 8, 10

WEIGHTING_FUNCS 2, 12
WEIGHTING_FUNCS 4, 12
WEIGHTING_FUNCS 6, 12
WEIGHTING_FUNCS 8, 12

HEVC_BI_PEL_PIXELS  2, 8
HEVC_BI_PEL_PIXELS  4, 8
HEVC_BI_PEL_PIXELS  6, 8
HEVC_BI_PEL_PIXELS  8, 8
HEVC_BI_PEL_PIXELS 12, 8
HEVC_BI_PEL_PIXELS 16, 8

HEVC_BI_PEL_PIXELS 2, 10
HEVC_BI_PEL_PIXELS 4, 10
HEVC_BI_PEL_PIXELS 6, 10
HEVC_BI_PEL_PIXELS 8, 10

HEVC_BI_PEL_PIXELS 2, 12
HEVC_BI_PEL_PIXELS 4, 12
HEVC_BI_PEL_PIXELS 6, 12
HEVC_BI_PEL_PIXELS 8, 12

HEVC_PUT_HEVC_EPEL 2,  8
HEVC_PUT_HEVC_EPEL 4,  8
HEVC_PUT_HEVC_EPEL 6,  8
HEVC_PUT_HEVC_EPEL 8,  8
HEVC_PUT_HEVC_EPEL 12, 8
HEVC_PUT_HEVC_EPEL 16, 8


HEVC_PUT_HEVC_EPEL 2, 10
HEVC_PUT_HEVC_EPEL 4, 10
HEVC_PUT_HEVC_EPEL 6, 10
HEVC_PUT_HEVC_EPEL 8, 10

HEVC_PUT_HEVC_EPEL 2, 12
HEVC_PUT_HEVC_EPEL 4, 12
HEVC_PUT_HEVC_EPEL 6, 12
HEVC_PUT_HEVC_EPEL 8, 12

HEVC_PUT_HEVC_EPEL_HV 2,  8
HEVC_PUT_HEVC_EPEL_HV 4,  8
HEVC_PUT_HEVC_EPEL_HV 6,  8
HEVC_PUT_HEVC_EPEL_HV 8,  8
HEVC_PUT_HEVC_EPEL_HV 16, 8

HEVC_PUT_HEVC_EPEL_HV 2, 10
HEVC_PUT_HEVC_EPEL_HV 4, 10
HEVC_PUT_HEVC_EPEL_HV 6, 10
HEVC_PUT_HEVC_EPEL_HV 8, 10

HEVC_PUT_HEVC_EPEL_HV 2, 12
HEVC_PUT_HEVC_EPEL_HV 4, 12
HEVC_PUT_HEVC_EPEL_HV 6, 12
HEVC_PUT_HEVC_EPEL_HV 8, 12

HEVC_PUT_HEVC_QPEL 4,  8
HEVC_PUT_HEVC_QPEL 8,  8
HEVC_PUT_HEVC_QPEL 12, 8
HEVC_PUT_HEVC_QPEL 16, 8

HEVC_PUT_HEVC_QPEL 4, 10
HEVC_PUT_HEVC_QPEL 8, 10

HEVC_PUT_HEVC_QPEL 4, 12
HEVC_PUT_HEVC_QPEL 8, 12

HEVC_PUT_HEVC_QPEL_HV 2, 8
HEVC_PUT_HEVC_QPEL_HV 4, 8
HEVC_PUT_HEVC_QPEL_HV 6, 8
HEVC_PUT_HEVC_QPEL_HV 8, 8

HEVC_PUT_HEVC_QPEL_HV 2, 10
HEVC_PUT_HEVC_QPEL_HV 4, 10
HEVC_PUT_HEVC_QPEL_HV 6, 10
HEVC_PUT_HEVC_QPEL_HV 8, 10

HEVC_PUT_HEVC_QPEL_HV 2, 12
HEVC_PUT_HEVC_QPEL_HV 4, 12
HEVC_PUT_HEVC_QPEL_HV 6, 12
HEVC_PUT_HEVC_QPEL_HV 8, 12

%if HAVE_AVX2_EXTERNAL
INIT_YMM avx2  ; adds ff_ and _avx2 to function name & enables 256b registers : m0 for 256b, xm0 for 128b. cpuflag(avx2) = 1 / notcpuflag(avx) = 0

HEVC_BI_PEL_PIXELS 32, 8
HEVC_BI_PEL_PIXELS 16, 10

HEVC_PUT_HEVC_EPEL 32, 8
HEVC_PUT_HEVC_EPEL 16, 10

HEVC_PUT_HEVC_EPEL_HV 16, 10
HEVC_PUT_HEVC_EPEL_HV 32, 8

HEVC_PUT_HEVC_QPEL 32, 8

HEVC_PUT_HEVC_QPEL 16, 10

HEVC_PUT_HEVC_QPEL_HV 16, 10

%endif ;AVX2
%endif ; ARCH_X86_64

%macro QPEL_FILTER_H 5
%define %%table hevc_qpel_filters_avx512icl_h_%1
%assign %%offset 4
    dec %2q
    shl %2q, 3
%if PIC
    lea %5q, [%%table]
    %define FILTER %5q
%else
    %define FILTER %%table
%endif
    vpbroadcastd m%3, [FILTER + %2q + 0*%%offset]
    vpbroadcastd m%4, [FILTER + %2q + 1*%%offset]
%endmacro

%macro QPEL_FILTER_V 5
    vpbroadcastd m%3, [%5 + %2q + 4*%4]
%endmacro

%macro QPEL_LOAD_SHUF 2
    movu m%1, [pb_qpel_shuffle_index +  0]
    movu m%2, [pb_qpel_shuffle_index + 64]
%endmacro

; required: m0-m5
; %1: dst register index
; %2: name for src
; %3: optional offset
%macro QPEL_H_LOAD_COMPUTE 2-3
%assign %%offset 0
%if %0 == 3
%assign %%offset %3
%endif
    pxor            m%1, m%1
%if mmsize == 64
    movu            ym4, [%2q + %%offset - 3]
%else
    movu            xm4, [%2q + %%offset - 3]
%endif
    vpermb           m5, m2, m4
    vpermb           m4, m3, m4
    vpdpbusd        m%1, m5, m0
    vpdpbusd        m%1, m4, m1
%endmacro

%macro HEVC_PUT_HEVC_QPEL_AVX512ICL 2
cglobal hevc_put_hevc_qpel_h%1_%2, 5, 6, 8, dst, src, srcstride, height, mx, tmp
    QPEL_FILTER_H   %1, mx, 0, 1, tmp
    QPEL_LOAD_SHUF   2, 3
.loop:
    QPEL_H_LOAD_COMPUTE   6, src
%if %1 == 4
    vpmovdw             xm6, m6
    movq             [dstq], xm6
%else
    vpmovdw          [dstq], m6
%endif
%if %1 > 16
    QPEL_H_LOAD_COMPUTE   7, src, 16
    vpmovdw     [dstq + 32], m7
%endif
%if %1 > 32
    QPEL_H_LOAD_COMPUTE   6, src, 32
    QPEL_H_LOAD_COMPUTE   7, src, 48
    vpmovdw     [dstq + 64], m6
    vpmovdw     [dstq + 96], m7
%endif
    LOOP_END            dst, src, srcstride
    RET
%endmacro

%macro HEVC_PUT_HEVC_QPEL_HV_AVX512ICL 2
cglobal hevc_put_hevc_qpel_hv%1_%2, 6, 7, 27, dst, src, srcstride, height, mx, my, tmp
%assign %%shift 6
%assign %%extra 7
    QPEL_FILTER_H    %1, mx, 0, 1, tmp
    QPEL_LOAD_SHUF    2, 3
    lea            tmpq, [srcstrideq*3]
    sub            srcq, tmpq
    sub             myq, 1
    shl             myq, 5
%define %%table hevc_qpel_filters_avx512icl_v_%1
%if PIC
    lea tmpq, [%%table]
    %define FILTER tmpq
%else
    %define FILTER %%table
%endif
%assign %%i 6
%assign %%j 0
%rep %1
    QPEL_FILTER_V %1, my, %%i, %%j, FILTER
    %assign %%i %%i+1
    %assign %%j %%j+1
%endrep
%rep %%extra
    QPEL_H_LOAD_COMPUTE %%i, src
    add srcq, srcstrideq
%assign %%i %%i+1
%endrep
.loop:
    QPEL_H_LOAD_COMPUTE %%i, src
    vpmulld           m22, m14, m6
    vpmulld           m23, m15, m7
    vpmulld           m24, m16, m8
    vpmulld           m25, m17, m9
    vpaddd            m26, m22, m23
    vpaddd            m24, m25
    vpaddd            m26, m24
    vpmulld           m22, m18, m10
    vpmulld           m23, m19, m11
    vpmulld           m24, m20, m12
    vpmulld           m25, m21, m13
    vpaddd            m22, m22, m23
    vpaddd            m24, m25
    vpaddd            m26, m24
    vpaddd            m22, m26
    mova              m14, m15
    mova              m15, m16
    mova              m16, m17
    mova              m17, m18
    mova              m18, m19
    mova              m19, m20
    mova              m20, m21
    vpsrad            m22, %%shift
    vpmovdw        [dstq], m22
    LOOP_END          dst, src, srcstride

    RET
%endmacro

%if ARCH_X86_64
%if HAVE_AVX512ICL_EXTERNAL

INIT_XMM avx512icl
HEVC_PUT_HEVC_QPEL_AVX512ICL 4, 8

INIT_YMM avx512icl
HEVC_PUT_HEVC_QPEL_AVX512ICL 8, 8
HEVC_PUT_HEVC_QPEL_HV_AVX512ICL 8, 8

INIT_ZMM avx512icl
HEVC_PUT_HEVC_QPEL_AVX512ICL 16, 8
HEVC_PUT_HEVC_QPEL_AVX512ICL 32, 8
HEVC_PUT_HEVC_QPEL_AVX512ICL 64, 8

%endif
%endif