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
|
;
; Copyright (c) 2010 The WebM 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 "vpx_ports/x86_abi_support.asm"
;void copy_mem16x16_sse2(
; unsigned char *src,
; int src_stride,
; unsigned char *dst,
; int dst_stride
; )
global sym(vp8_copy_mem16x16_sse2) PRIVATE
sym(vp8_copy_mem16x16_sse2):
push rbp
mov rbp, rsp
SHADOW_ARGS_TO_STACK 4
push rsi
push rdi
; end prolog
mov rsi, arg(0) ;src;
movdqu xmm0, [rsi]
movsxd rax, dword ptr arg(1) ;src_stride;
mov rdi, arg(2) ;dst;
movdqu xmm1, [rsi+rax]
movdqu xmm2, [rsi+rax*2]
movsxd rcx, dword ptr arg(3) ;dst_stride
lea rsi, [rsi+rax*2]
movdqa [rdi], xmm0
add rsi, rax
movdqa [rdi+rcx], xmm1
movdqa [rdi+rcx*2],xmm2
lea rdi, [rdi+rcx*2]
movdqu xmm3, [rsi]
add rdi, rcx
movdqu xmm4, [rsi+rax]
movdqu xmm5, [rsi+rax*2]
lea rsi, [rsi+rax*2]
movdqa [rdi], xmm3
add rsi, rax
movdqa [rdi+rcx], xmm4
movdqa [rdi+rcx*2],xmm5
lea rdi, [rdi+rcx*2]
movdqu xmm0, [rsi]
add rdi, rcx
movdqu xmm1, [rsi+rax]
movdqu xmm2, [rsi+rax*2]
lea rsi, [rsi+rax*2]
movdqa [rdi], xmm0
add rsi, rax
movdqa [rdi+rcx], xmm1
movdqa [rdi+rcx*2], xmm2
movdqu xmm3, [rsi]
movdqu xmm4, [rsi+rax]
lea rdi, [rdi+rcx*2]
add rdi, rcx
movdqu xmm5, [rsi+rax*2]
lea rsi, [rsi+rax*2]
movdqa [rdi], xmm3
add rsi, rax
movdqa [rdi+rcx], xmm4
movdqa [rdi+rcx*2],xmm5
movdqu xmm0, [rsi]
lea rdi, [rdi+rcx*2]
movdqu xmm1, [rsi+rax]
add rdi, rcx
movdqu xmm2, [rsi+rax*2]
lea rsi, [rsi+rax*2]
movdqa [rdi], xmm0
movdqa [rdi+rcx], xmm1
movdqa [rdi+rcx*2],xmm2
movdqu xmm3, [rsi+rax]
lea rdi, [rdi+rcx*2]
movdqa [rdi+rcx], xmm3
; begin epilog
pop rdi
pop rsi
UNSHADOW_ARGS
pop rbp
ret
;void vp8_intra_pred_uv_dc_mmx2(
; unsigned char *dst,
; int dst_stride
; unsigned char *above,
; unsigned char *left,
; int left_stride,
; )
global sym(vp8_intra_pred_uv_dc_mmx2) PRIVATE
sym(vp8_intra_pred_uv_dc_mmx2):
push rbp
mov rbp, rsp
SHADOW_ARGS_TO_STACK 5
push rsi
push rdi
; end prolog
; from top
mov rdi, arg(2) ;above;
mov rsi, arg(3) ;left;
movsxd rax, dword ptr arg(4) ;left_stride;
pxor mm0, mm0
movq mm1, [rdi]
lea rdi, [rax*3]
psadbw mm1, mm0
; from left
movzx ecx, byte [rsi]
movzx edx, byte [rsi+rax*1]
add ecx, edx
movzx edx, byte [rsi+rax*2]
add ecx, edx
movzx edx, byte [rsi+rdi]
lea rsi, [rsi+rax*4]
add ecx, edx
movzx edx, byte [rsi]
add ecx, edx
movzx edx, byte [rsi+rax]
add ecx, edx
movzx edx, byte [rsi+rax*2]
add ecx, edx
movzx edx, byte [rsi+rdi]
add ecx, edx
; add up
pextrw edx, mm1, 0x0
lea edx, [edx+ecx+8]
sar edx, 4
movd mm1, edx
movsxd rcx, dword ptr arg(1) ;dst_stride
pshufw mm1, mm1, 0x0
mov rdi, arg(0) ;dst;
packuswb mm1, mm1
; write out
lea rax, [rcx*3]
lea rdx, [rdi+rcx*4]
movq [rdi ], mm1
movq [rdi+rcx ], mm1
movq [rdi+rcx*2], mm1
movq [rdi+rax ], mm1
movq [rdx ], mm1
movq [rdx+rcx ], mm1
movq [rdx+rcx*2], mm1
movq [rdx+rax ], mm1
; begin epilog
pop rdi
pop rsi
UNSHADOW_ARGS
pop rbp
ret
;void vp8_intra_pred_uv_dctop_mmx2(
; unsigned char *dst,
; int dst_stride
; unsigned char *above,
; unsigned char *left,
; int left_stride,
; )
global sym(vp8_intra_pred_uv_dctop_mmx2) PRIVATE
sym(vp8_intra_pred_uv_dctop_mmx2):
push rbp
mov rbp, rsp
SHADOW_ARGS_TO_STACK 5
GET_GOT rbx
push rsi
push rdi
; end prolog
;arg(3), arg(4) not used
; from top
mov rsi, arg(2) ;above;
pxor mm0, mm0
movq mm1, [rsi]
psadbw mm1, mm0
; add up
paddw mm1, [GLOBAL(dc_4)]
psraw mm1, 3
pshufw mm1, mm1, 0x0
packuswb mm1, mm1
; write out
mov rdi, arg(0) ;dst;
movsxd rcx, dword ptr arg(1) ;dst_stride
lea rax, [rcx*3]
movq [rdi ], mm1
movq [rdi+rcx ], mm1
movq [rdi+rcx*2], mm1
movq [rdi+rax ], mm1
lea rdi, [rdi+rcx*4]
movq [rdi ], mm1
movq [rdi+rcx ], mm1
movq [rdi+rcx*2], mm1
movq [rdi+rax ], mm1
; begin epilog
pop rdi
pop rsi
RESTORE_GOT
UNSHADOW_ARGS
pop rbp
ret
;void vp8_intra_pred_uv_dcleft_mmx2(
; unsigned char *dst,
; int dst_stride
; unsigned char *above,
; unsigned char *left,
; int left_stride,
; )
global sym(vp8_intra_pred_uv_dcleft_mmx2) PRIVATE
sym(vp8_intra_pred_uv_dcleft_mmx2):
push rbp
mov rbp, rsp
SHADOW_ARGS_TO_STACK 5
push rsi
push rdi
; end prolog
;arg(2) not used
; from left
mov rsi, arg(3) ;left;
movsxd rax, dword ptr arg(4) ;left_stride;
lea rdi, [rax*3]
movzx ecx, byte [rsi]
movzx edx, byte [rsi+rax]
add ecx, edx
movzx edx, byte [rsi+rax*2]
add ecx, edx
movzx edx, byte [rsi+rdi]
add ecx, edx
lea rsi, [rsi+rax*4]
movzx edx, byte [rsi]
add ecx, edx
movzx edx, byte [rsi+rax]
add ecx, edx
movzx edx, byte [rsi+rax*2]
add ecx, edx
movzx edx, byte [rsi+rdi]
lea edx, [ecx+edx+4]
; add up
shr edx, 3
movd mm1, edx
pshufw mm1, mm1, 0x0
packuswb mm1, mm1
; write out
mov rdi, arg(0) ;dst;
movsxd rcx, dword ptr arg(1) ;dst_stride
lea rax, [rcx*3]
movq [rdi ], mm1
movq [rdi+rcx ], mm1
movq [rdi+rcx*2], mm1
movq [rdi+rax ], mm1
lea rdi, [rdi+rcx*4]
movq [rdi ], mm1
movq [rdi+rcx ], mm1
movq [rdi+rcx*2], mm1
movq [rdi+rax ], mm1
; begin epilog
pop rdi
pop rsi
UNSHADOW_ARGS
pop rbp
ret
;void vp8_intra_pred_uv_dc128_mmx(
; unsigned char *dst,
; int dst_stride
; unsigned char *above,
; unsigned char *left,
; int left_stride,
; )
global sym(vp8_intra_pred_uv_dc128_mmx) PRIVATE
sym(vp8_intra_pred_uv_dc128_mmx):
push rbp
mov rbp, rsp
SHADOW_ARGS_TO_STACK 5
GET_GOT rbx
; end prolog
;arg(2), arg(3), arg(4) not used
; write out
movq mm1, [GLOBAL(dc_128)]
mov rax, arg(0) ;dst;
movsxd rdx, dword ptr arg(1) ;dst_stride
lea rcx, [rdx*3]
movq [rax ], mm1
movq [rax+rdx ], mm1
movq [rax+rdx*2], mm1
movq [rax+rcx ], mm1
lea rax, [rax+rdx*4]
movq [rax ], mm1
movq [rax+rdx ], mm1
movq [rax+rdx*2], mm1
movq [rax+rcx ], mm1
; begin epilog
RESTORE_GOT
UNSHADOW_ARGS
pop rbp
ret
;void vp8_intra_pred_uv_tm_sse2(
; unsigned char *dst,
; int dst_stride
; unsigned char *above,
; unsigned char *left,
; int left_stride,
; )
%macro vp8_intra_pred_uv_tm 1
global sym(vp8_intra_pred_uv_tm_%1) PRIVATE
sym(vp8_intra_pred_uv_tm_%1):
push rbp
mov rbp, rsp
SHADOW_ARGS_TO_STACK 5
GET_GOT rbx
push rsi
push rdi
push rbx
; end prolog
; read top row
mov edx, 4
mov rsi, arg(2) ;above
movsxd rax, dword ptr arg(4) ;left_stride;
pxor xmm0, xmm0
%ifidn %1, ssse3
movdqa xmm2, [GLOBAL(dc_1024)]
%endif
movq xmm1, [rsi]
punpcklbw xmm1, xmm0
; set up left ptrs ans subtract topleft
movd xmm3, [rsi-1]
mov rsi, arg(3) ;left;
%ifidn %1, sse2
punpcklbw xmm3, xmm0
pshuflw xmm3, xmm3, 0x0
punpcklqdq xmm3, xmm3
%else
pshufb xmm3, xmm2
%endif
psubw xmm1, xmm3
; set up dest ptrs
mov rdi, arg(0) ;dst;
movsxd rcx, dword ptr arg(1) ;dst_stride
.vp8_intra_pred_uv_tm_%1_loop:
mov bl, [rsi]
movd xmm3, ebx
mov bl, [rsi+rax]
movd xmm5, ebx
%ifidn %1, sse2
punpcklbw xmm3, xmm0
punpcklbw xmm5, xmm0
pshuflw xmm3, xmm3, 0x0
pshuflw xmm5, xmm5, 0x0
punpcklqdq xmm3, xmm3
punpcklqdq xmm5, xmm5
%else
pshufb xmm3, xmm2
pshufb xmm5, xmm2
%endif
paddw xmm3, xmm1
paddw xmm5, xmm1
packuswb xmm3, xmm5
movq [rdi ], xmm3
movhps[rdi+rcx], xmm3
lea rsi, [rsi+rax*2]
lea rdi, [rdi+rcx*2]
dec edx
jnz .vp8_intra_pred_uv_tm_%1_loop
; begin epilog
pop rbx
pop rdi
pop rsi
RESTORE_GOT
UNSHADOW_ARGS
pop rbp
ret
%endmacro
vp8_intra_pred_uv_tm sse2
vp8_intra_pred_uv_tm ssse3
;void vp8_intra_pred_uv_ve_mmx(
; unsigned char *dst,
; int dst_stride
; unsigned char *above,
; unsigned char *left,
; int left_stride,
; )
global sym(vp8_intra_pred_uv_ve_mmx) PRIVATE
sym(vp8_intra_pred_uv_ve_mmx):
push rbp
mov rbp, rsp
SHADOW_ARGS_TO_STACK 5
; end prolog
; arg(3), arg(4) not used
; read from top
mov rax, arg(2) ;src;
movq mm1, [rax]
; write out
mov rax, arg(0) ;dst;
movsxd rdx, dword ptr arg(1) ;dst_stride
lea rcx, [rdx*3]
movq [rax ], mm1
movq [rax+rdx ], mm1
movq [rax+rdx*2], mm1
movq [rax+rcx ], mm1
lea rax, [rax+rdx*4]
movq [rax ], mm1
movq [rax+rdx ], mm1
movq [rax+rdx*2], mm1
movq [rax+rcx ], mm1
; begin epilog
UNSHADOW_ARGS
pop rbp
ret
;void vp8_intra_pred_uv_ho_mmx2(
; unsigned char *dst,
; int dst_stride
; unsigned char *above,
; unsigned char *left,
; int left_stride
; )
%macro vp8_intra_pred_uv_ho 1
global sym(vp8_intra_pred_uv_ho_%1) PRIVATE
sym(vp8_intra_pred_uv_ho_%1):
push rbp
mov rbp, rsp
SHADOW_ARGS_TO_STACK 5
push rsi
push rdi
push rbx
%ifidn %1, ssse3
GET_GOT rbx
%endif
; end prolog
;arg(2) not used
; read from left and write out
%ifidn %1, mmx2
mov edx, 4
%endif
mov rsi, arg(3) ;left
movsxd rax, dword ptr arg(4) ;left_stride;
mov rdi, arg(0) ;dst;
movsxd rcx, dword ptr arg(1) ;dst_stride
%ifidn %1, ssse3
lea rdx, [rcx*3]
movdqa xmm2, [GLOBAL(dc_00001111)]
%endif
%ifidn %1, mmx2
.vp8_intra_pred_uv_ho_%1_loop:
mov bl, [rsi]
movd mm0, ebx
mov bl, [rsi+rax]
movd mm1, ebx
punpcklbw mm0, mm0
punpcklbw mm1, mm1
pshufw mm0, mm0, 0x0
pshufw mm1, mm1, 0x0
movq [rdi ], mm0
movq [rdi+rcx], mm1
lea rsi, [rsi+rax*2]
lea rdi, [rdi+rcx*2]
dec edx
jnz .vp8_intra_pred_uv_ho_%1_loop
%else
mov bl, [rsi]
movd xmm0, ebx
mov bl, [rsi+rax]
movd xmm3, ebx
mov bl, [rsi+rax*2]
movd xmm1, ebx
lea rbx, [rax*3]
mov bl, [rsi+rbx]
movd xmm4, ebx
punpcklbw xmm0, xmm3
punpcklbw xmm1, xmm4
pshufb xmm0, xmm2
pshufb xmm1, xmm2
movq [rdi ], xmm0
movhps [rdi+rcx], xmm0
movq [rdi+rcx*2], xmm1
movhps [rdi+rdx], xmm1
lea rsi, [rsi+rax*4]
lea rdi, [rdi+rcx*4]
mov bl, [rsi]
movd xmm0, ebx
mov bl, [rsi+rax]
movd xmm3, ebx
mov bl, [rsi+rax*2]
movd xmm1, ebx
lea rbx, [rax*3]
mov bl, [rsi+rbx]
movd xmm4, ebx
punpcklbw xmm0, xmm3
punpcklbw xmm1, xmm4
pshufb xmm0, xmm2
pshufb xmm1, xmm2
movq [rdi ], xmm0
movhps [rdi+rcx], xmm0
movq [rdi+rcx*2], xmm1
movhps [rdi+rdx], xmm1
%endif
; begin epilog
%ifidn %1, ssse3
RESTORE_GOT
%endif
pop rbx
pop rdi
pop rsi
UNSHADOW_ARGS
pop rbp
ret
%endmacro
vp8_intra_pred_uv_ho mmx2
vp8_intra_pred_uv_ho ssse3
;void vp8_intra_pred_y_dc_sse2(
; unsigned char *dst,
; int dst_stride
; unsigned char *above,
; unsigned char *left,
; int left_stride
; )
global sym(vp8_intra_pred_y_dc_sse2) PRIVATE
sym(vp8_intra_pred_y_dc_sse2):
push rbp
mov rbp, rsp
SHADOW_ARGS_TO_STACK 5
push rsi
push rdi
; end prolog
; from top
mov rdi, arg(2) ;above
mov rsi, arg(3) ;left
movsxd rax, dword ptr arg(4) ;left_stride;
pxor xmm0, xmm0
movdqa xmm1, [rdi]
psadbw xmm1, xmm0
movq xmm2, xmm1
punpckhqdq xmm1, xmm1
paddw xmm1, xmm2
; from left
lea rdi, [rax*3]
movzx ecx, byte [rsi]
movzx edx, byte [rsi+rax]
add ecx, edx
movzx edx, byte [rsi+rax*2]
add ecx, edx
movzx edx, byte [rsi+rdi]
add ecx, edx
lea rsi, [rsi+rax*4]
movzx edx, byte [rsi]
add ecx, edx
movzx edx, byte [rsi+rax]
add ecx, edx
movzx edx, byte [rsi+rax*2]
add ecx, edx
movzx edx, byte [rsi+rdi]
add ecx, edx
lea rsi, [rsi+rax*4]
movzx edx, byte [rsi]
add ecx, edx
movzx edx, byte [rsi+rax]
add ecx, edx
movzx edx, byte [rsi+rax*2]
add ecx, edx
movzx edx, byte [rsi+rdi]
add ecx, edx
lea rsi, [rsi+rax*4]
movzx edx, byte [rsi]
add ecx, edx
movzx edx, byte [rsi+rax]
add ecx, edx
movzx edx, byte [rsi+rax*2]
add ecx, edx
movzx edx, byte [rsi+rdi]
add ecx, edx
; add up
pextrw edx, xmm1, 0x0
lea edx, [edx+ecx+16]
sar edx, 5
movd xmm1, edx
; FIXME use pshufb for ssse3 version
pshuflw xmm1, xmm1, 0x0
punpcklqdq xmm1, xmm1
packuswb xmm1, xmm1
; write out
mov rsi, 2
mov rdi, arg(0) ;dst;
movsxd rcx, dword ptr arg(1) ;dst_stride
lea rax, [rcx*3]
.label
movdqa [rdi ], xmm1
movdqa [rdi+rcx ], xmm1
movdqa [rdi+rcx*2], xmm1
movdqa [rdi+rax ], xmm1
lea rdi, [rdi+rcx*4]
movdqa [rdi ], xmm1
movdqa [rdi+rcx ], xmm1
movdqa [rdi+rcx*2], xmm1
movdqa [rdi+rax ], xmm1
lea rdi, [rdi+rcx*4]
dec rsi
jnz .label
; begin epilog
pop rdi
pop rsi
UNSHADOW_ARGS
pop rbp
ret
;void vp8_intra_pred_y_dctop_sse2(
; unsigned char *dst,
; int dst_stride
; unsigned char *above,
; unsigned char *left,
; int left_stride
; )
global sym(vp8_intra_pred_y_dctop_sse2) PRIVATE
sym(vp8_intra_pred_y_dctop_sse2):
push rbp
mov rbp, rsp
SHADOW_ARGS_TO_STACK 5
push rsi
GET_GOT rbx
; end prolog
;arg(3), arg(4) not used
; from top
mov rcx, arg(2) ;above;
pxor xmm0, xmm0
movdqa xmm1, [rcx]
psadbw xmm1, xmm0
movdqa xmm2, xmm1
punpckhqdq xmm1, xmm1
paddw xmm1, xmm2
; add up
paddw xmm1, [GLOBAL(dc_8)]
psraw xmm1, 4
; FIXME use pshufb for ssse3 version
pshuflw xmm1, xmm1, 0x0
punpcklqdq xmm1, xmm1
packuswb xmm1, xmm1
; write out
mov rsi, 2
mov rdx, arg(0) ;dst;
movsxd rcx, dword ptr arg(1) ;dst_stride
lea rax, [rcx*3]
.label
movdqa [rdx ], xmm1
movdqa [rdx+rcx ], xmm1
movdqa [rdx+rcx*2], xmm1
movdqa [rdx+rax ], xmm1
lea rdx, [rdx+rcx*4]
movdqa [rdx ], xmm1
movdqa [rdx+rcx ], xmm1
movdqa [rdx+rcx*2], xmm1
movdqa [rdx+rax ], xmm1
lea rdx, [rdx+rcx*4]
dec rsi
jnz .label
; begin epilog
RESTORE_GOT
pop rsi
UNSHADOW_ARGS
pop rbp
ret
;void vp8_intra_pred_y_dcleft_sse2(
; unsigned char *dst,
; int dst_stride
; unsigned char *above,
; unsigned char *left,
; int left_stride
; )
global sym(vp8_intra_pred_y_dcleft_sse2) PRIVATE
sym(vp8_intra_pred_y_dcleft_sse2):
push rbp
mov rbp, rsp
SHADOW_ARGS_TO_STACK 5
push rsi
push rdi
; end prolog
;arg(2) not used
; from left
mov rsi, arg(3) ;left;
movsxd rax, dword ptr arg(4) ;left_stride;
lea rdi, [rax*3]
movzx ecx, byte [rsi]
movzx edx, byte [rsi+rax]
add ecx, edx
movzx edx, byte [rsi+rax*2]
add ecx, edx
movzx edx, byte [rsi+rdi]
add ecx, edx
lea rsi, [rsi+rax*4]
movzx edx, byte [rsi]
add ecx, edx
movzx edx, byte [rsi+rax]
add ecx, edx
movzx edx, byte [rsi+rax*2]
add ecx, edx
movzx edx, byte [rsi+rdi]
add ecx, edx
lea rsi, [rsi+rax*4]
movzx edx, byte [rsi]
add ecx, edx
movzx edx, byte [rsi+rax]
add ecx, edx
movzx edx, byte [rsi+rax*2]
add ecx, edx
movzx edx, byte [rsi+rdi]
add ecx, edx
lea rsi, [rsi+rax*4]
movzx edx, byte [rsi]
add ecx, edx
movzx edx, byte [rsi+rax]
add ecx, edx
movzx edx, byte [rsi+rax*2]
add ecx, edx
movzx edx, byte [rsi+rdi]
lea edx, [ecx+edx+8]
; add up
shr edx, 4
movd xmm1, edx
; FIXME use pshufb for ssse3 version
pshuflw xmm1, xmm1, 0x0
punpcklqdq xmm1, xmm1
packuswb xmm1, xmm1
; write out
mov rsi, 2
mov rdi, arg(0) ;dst;
movsxd rcx, dword ptr arg(1) ;dst_stride
lea rax, [rcx*3]
.label
movdqa [rdi ], xmm1
movdqa [rdi+rcx ], xmm1
movdqa [rdi+rcx*2], xmm1
movdqa [rdi+rax ], xmm1
lea rdi, [rdi+rcx*4]
movdqa [rdi ], xmm1
movdqa [rdi+rcx ], xmm1
movdqa [rdi+rcx*2], xmm1
movdqa [rdi+rax ], xmm1
lea rdi, [rdi+rcx*4]
dec rsi
jnz .label
; begin epilog
pop rdi
pop rsi
UNSHADOW_ARGS
pop rbp
ret
;void vp8_intra_pred_y_dc128_sse2(
; unsigned char *dst,
; int dst_stride
; unsigned char *above,
; unsigned char *left,
; int left_stride
; )
global sym(vp8_intra_pred_y_dc128_sse2) PRIVATE
sym(vp8_intra_pred_y_dc128_sse2):
push rbp
mov rbp, rsp
SHADOW_ARGS_TO_STACK 5
push rsi
GET_GOT rbx
; end prolog
;arg(2), arg(3), arg(4) not used
; write out
mov rsi, 2
movdqa xmm1, [GLOBAL(dc_128)]
mov rax, arg(0) ;dst;
movsxd rdx, dword ptr arg(1) ;dst_stride
lea rcx, [rdx*3]
.label
movdqa [rax ], xmm1
movdqa [rax+rdx ], xmm1
movdqa [rax+rdx*2], xmm1
movdqa [rax+rcx ], xmm1
lea rax, [rax+rdx*4]
movdqa [rax ], xmm1
movdqa [rax+rdx ], xmm1
movdqa [rax+rdx*2], xmm1
movdqa [rax+rcx ], xmm1
lea rax, [rax+rdx*4]
dec rsi
jnz .label
; begin epilog
RESTORE_GOT
pop rsi
UNSHADOW_ARGS
pop rbp
ret
;void vp8_intra_pred_y_tm_sse2(
; unsigned char *dst,
; int dst_stride
; unsigned char *above,
; unsigned char *left,
; int left_stride
; )
%macro vp8_intra_pred_y_tm 1
global sym(vp8_intra_pred_y_tm_%1) PRIVATE
sym(vp8_intra_pred_y_tm_%1):
push rbp
mov rbp, rsp
SHADOW_ARGS_TO_STACK 5
SAVE_XMM 7
push rsi
push rdi
push rbx
GET_GOT rbx
; end prolog
; read top row
mov edx, 8
mov rsi, arg(2) ;above
movsxd rax, dword ptr arg(4) ;left_stride;
pxor xmm0, xmm0
%ifidn %1, ssse3
movdqa xmm3, [GLOBAL(dc_1024)]
%endif
movdqa xmm1, [rsi]
movdqa xmm2, xmm1
punpcklbw xmm1, xmm0
punpckhbw xmm2, xmm0
; set up left ptrs ans subtract topleft
movd xmm4, [rsi-1]
mov rsi, arg(3) ;left
%ifidn %1, sse2
punpcklbw xmm4, xmm0
pshuflw xmm4, xmm4, 0x0
punpcklqdq xmm4, xmm4
%else
pshufb xmm4, xmm3
%endif
psubw xmm1, xmm4
psubw xmm2, xmm4
; set up dest ptrs
mov rdi, arg(0) ;dst;
movsxd rcx, dword ptr arg(1) ;dst_stride
vp8_intra_pred_y_tm_%1_loop:
mov bl, [rsi]
movd xmm4, ebx
mov bl, [rsi+rax]
movd xmm5, ebx
%ifidn %1, sse2
punpcklbw xmm4, xmm0
punpcklbw xmm5, xmm0
pshuflw xmm4, xmm4, 0x0
pshuflw xmm5, xmm5, 0x0
punpcklqdq xmm4, xmm4
punpcklqdq xmm5, xmm5
%else
pshufb xmm4, xmm3
pshufb xmm5, xmm3
%endif
movdqa xmm6, xmm4
movdqa xmm7, xmm5
paddw xmm4, xmm1
paddw xmm6, xmm2
paddw xmm5, xmm1
paddw xmm7, xmm2
packuswb xmm4, xmm6
packuswb xmm5, xmm7
movdqa [rdi ], xmm4
movdqa [rdi+rcx], xmm5
lea rsi, [rsi+rax*2]
lea rdi, [rdi+rcx*2]
dec edx
jnz vp8_intra_pred_y_tm_%1_loop
; begin epilog
RESTORE_GOT
pop rbx
pop rdi
pop rsi
RESTORE_XMM
UNSHADOW_ARGS
pop rbp
ret
%endmacro
vp8_intra_pred_y_tm sse2
vp8_intra_pred_y_tm ssse3
;void vp8_intra_pred_y_ve_sse2(
; unsigned char *dst,
; int dst_stride
; unsigned char *above,
; unsigned char *left,
; int left_stride
; )
global sym(vp8_intra_pred_y_ve_sse2) PRIVATE
sym(vp8_intra_pred_y_ve_sse2):
push rbp
mov rbp, rsp
SHADOW_ARGS_TO_STACK 5
push rsi
; end prolog
;arg(3), arg(4) not used
mov rax, arg(2) ;above;
mov rsi, 2
movsxd rdx, dword ptr arg(1) ;dst_stride
; read from top
movdqa xmm1, [rax]
; write out
mov rax, arg(0) ;dst;
lea rcx, [rdx*3]
.label
movdqa [rax ], xmm1
movdqa [rax+rdx ], xmm1
movdqa [rax+rdx*2], xmm1
movdqa [rax+rcx ], xmm1
lea rax, [rax+rdx*4]
movdqa [rax ], xmm1
movdqa [rax+rdx ], xmm1
movdqa [rax+rdx*2], xmm1
movdqa [rax+rcx ], xmm1
lea rax, [rax+rdx*4]
dec rsi
jnz .label
; begin epilog
pop rsi
UNSHADOW_ARGS
pop rbp
ret
;void vp8_intra_pred_y_ho_sse2(
; unsigned char *dst,
; int dst_stride
; unsigned char *above,
; unsigned char *left,
; int left_stride,
; )
global sym(vp8_intra_pred_y_ho_sse2) PRIVATE
sym(vp8_intra_pred_y_ho_sse2):
push rbp
mov rbp, rsp
SHADOW_ARGS_TO_STACK 5
push rsi
push rdi
push rbx
; end prolog
;arg(2) not used
; read from left and write out
mov edx, 8
mov rsi, arg(3) ;left;
movsxd rax, dword ptr arg(4) ;left_stride;
mov rdi, arg(0) ;dst;
movsxd rcx, dword ptr arg(1) ;dst_stride
vp8_intra_pred_y_ho_sse2_loop:
mov bl, [rsi]
movd xmm0, ebx
mov bl, [rsi+rax]
movd xmm1, ebx
; FIXME use pshufb for ssse3 version
punpcklbw xmm0, xmm0
punpcklbw xmm1, xmm1
pshuflw xmm0, xmm0, 0x0
pshuflw xmm1, xmm1, 0x0
punpcklqdq xmm0, xmm0
punpcklqdq xmm1, xmm1
movdqa [rdi ], xmm0
movdqa [rdi+rcx], xmm1
lea rsi, [rsi+rax*2]
lea rdi, [rdi+rcx*2]
dec edx
jnz vp8_intra_pred_y_ho_sse2_loop
; begin epilog
pop rbx
pop rdi
pop rsi
UNSHADOW_ARGS
pop rbp
ret
SECTION_RODATA
align 16
dc_128:
times 16 db 128
dc_4:
times 4 dw 4
align 16
dc_8:
times 8 dw 8
align 16
dc_1024:
times 8 dw 0x400
align 16
dc_00001111:
times 8 db 0
times 8 db 1
|
