forked from EricLengyel/Terathon-Math-Library
-
Notifications
You must be signed in to change notification settings - Fork 0
/
TSSimd.h
2514 lines (1645 loc) · 56.7 KB
/
TSSimd.h
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
//
// This file is part of the Terathon Common Library, by Eric Lengyel.
// Copyright 1999-2023, Terathon Software LLC
//
// This software is distributed under the MIT License.
// Separate proprietary licenses are available from Terathon Software.
//
#ifndef TSSimd_h
#define TSSimd_h
#include "TSPlatform.h"
#if defined(_M_X64) || defined(_M_IX86) || defined(__x86_64__) || defined(__SSE__)
#define TERATHON_SSE 1
#endif
#if defined(__AVX__)
#define TERATHON_AVX 1
#endif
#if defined(__ARM_NEON)
#define TERATHON_NEON 1
#endif
#ifdef TERATHON_NO_SYSTEM
#if defined(TERATHON_SSE)
typedef union __declspec(intrin_type) __declspec(align(16)) __m128
{
float m128_f32[4];
unsigned __int64 m128_u64[2];
__int8 m128_i8[16];
__int16 m128_i16[8];
__int32 m128_i32[4];
__int64 m128_i64[2];
unsigned __int8 m128_u8[16];
unsigned __int16 m128_u16[8];
unsigned __int32 m128_u32[4];
} __m128;
typedef union __declspec(intrin_type) __declspec(align(16)) __m128i
{
__int8 m128i_i8[16];
__int16 m128i_i16[8];
__int32 m128i_i32[4];
__int64 m128i_i64[2];
unsigned __int8 m128i_u8[16];
unsigned __int16 m128i_u16[8];
unsigned __int32 m128i_u32[4];
unsigned __int64 m128i_u64[2];
} __m128i;
#define _MM_SHUFFLE(p3,p2,p1,p0) (((p3) << 6) | ((p2) << 4) | ((p1) << 2) | (p0))
extern "C"
{
extern __m128 _mm_add_ss(__m128, __m128);
extern __m128 _mm_add_ps(__m128, __m128);
extern __m128 _mm_sub_ss(__m128, __m128);
extern __m128 _mm_sub_ps(__m128, __m128);
extern __m128 _mm_mul_ss(__m128, __m128);
extern __m128 _mm_mul_ps(__m128, __m128);
extern __m128 _mm_div_ss(__m128, __m128);
extern __m128 _mm_div_ps(__m128, __m128);
extern __m128 _mm_rsqrt_ss(__m128);
extern __m128 _mm_rsqrt_ps(__m128);
extern __m128 _mm_min_ss(__m128, __m128);
extern __m128 _mm_min_ps(__m128, __m128);
extern __m128 _mm_max_ss(__m128, __m128);
extern __m128 _mm_max_ps(__m128, __m128);
extern __m128 _mm_and_ps(__m128, __m128);
extern __m128 _mm_andnot_ps(__m128, __m128);
extern __m128 _mm_or_ps(__m128, __m128);
extern __m128 _mm_xor_ps(__m128, __m128);
extern __m128 _mm_cmpeq_ps(__m128, __m128);
extern __m128 _mm_cmplt_ss(__m128, __m128);
extern __m128 _mm_cmplt_ps(__m128, __m128);
extern int _mm_comieq_ss(__m128, __m128);
extern int _mm_comilt_ss(__m128, __m128);
extern int _mm_comigt_ss(__m128, __m128);
extern int _mm_cvtt_ss2si(__m128);
extern __m128 _mm_shuffle_ps(__m128, __m128, unsigned int);
extern __m128 _mm_setzero_ps(void);
extern __m128 _mm_load_ss(const float *);
extern __m128 _mm_load_ps(const float *);
extern __m128 _mm_loadu_ps(const float *);
extern void _mm_store_ss(float *, __m128);
extern void _mm_store_ps(float *, __m128);
extern void _mm_storeu_ps(float *, __m128);
extern unsigned int _mm_getcsr(void);
extern void _mm_setcsr(unsigned int);
extern __m128i _mm_setzero_si128(void);
extern __m128i _mm_load_si128(const __m128i *);
extern __m128i _mm_loadu_si128(const __m128i *);
extern void _mm_store_si128(__m128i *, __m128i);
extern void _mm_storeu_si128(__m128i *, __m128i);
extern int _mm_cvtsi128_si32(__m128i);
extern __m128i _mm_shuffle_epi32(__m128i, int);
extern __m128i _mm_shufflehi_epi16(__m128i, int);
extern __m128i _mm_shufflelo_epi16(__m128i, int);
extern __m128i _mm_packs_epi16(__m128i, __m128i);
extern __m128i _mm_packs_epi32(__m128i, __m128i);
extern __m128i _mm_packus_epi16(__m128i, __m128i);
extern __m128i _mm_unpacklo_epi8(__m128i, __m128i);
extern __m128i _mm_unpackhi_epi8(__m128i, __m128i);
extern __m128i _mm_unpacklo_epi16(__m128i, __m128i);
extern __m128i _mm_unpackhi_epi16(__m128i, __m128i);
extern __m128i _mm_cmplt_epi8(__m128i, __m128i);
extern __m128i _mm_cmplt_epi16(__m128i, __m128i);
extern __m128 _mm_cvtepi32_ps(__m128i);
extern __m128i _mm_cvtps_epi32(__m128);
extern __m128i _mm_add_epi32(__m128i, __m128i);
extern __m128i _mm_sub_epi32(__m128i, __m128i);
}
#endif
#if defined(TERATHON_AVX)
typedef union __declspec(intrin_type) __declspec(align(32)) __m256
{
float m256_f32[8];
} __m256;
typedef union __declspec(intrin_type) __declspec(align(32)) __m256i
{
__int8 m256i_i8[32];
__int16 m256i_i16[16];
__int32 m256i_i32[8];
__int64 m256i_i64[4];
unsigned __int8 m256i_u8[32];
unsigned __int16 m256i_u16[16];
unsigned __int32 m256i_u32[8];
unsigned __int64 m256i_u64[4];
} __m256i;
extern "C"
{
extern __m256 __cdecl _mm256_add_ps(__m256, __m256);
extern __m256 __cdecl _mm256_sub_ps(__m256, __m256);
extern __m256 __cdecl _mm256_mul_ps(__m256, __m256);
extern __m256 __cdecl _mm256_div_ps(__m256, __m256);
extern __m256 __cdecl _mm256_rsqrt_ps(__m256);
extern __m256 __cdecl _mm256_min_ps(__m256, __m256);
extern __m256 __cdecl _mm256_max_ps(__m256, __m256);
extern __m256 __cdecl _mm256_and_ps(__m256, __m256);
extern __m256 __cdecl _mm256_andnot_ps(__m256, __m256);
extern __m256 __cdecl _mm256_or_ps(__m256, __m256);
extern __m256 __cdecl _mm256_xor_ps(__m256, __m256);
extern __m256 __cdecl _mm256_cmp_ps(__m256, __m256, int);
extern __m256 __cdecl _mm256_permute_ps(__m256, int);
extern __m256 __cdecl _mm256_permute2f128_ps(__m256, __m256, int);
extern __m256 __cdecl _mm256_insertf128_ps(__m256, __m128, int);
extern __m128 __cdecl _mm256_castps256_ps128(__m256);
extern __m256 __cdecl _mm256_castps128_ps256(__m128);
extern __m256 __cdecl _mm256_setzero_ps(void);
extern __m256 __cdecl _mm256_load_ps(const float *);
extern __m256 __cdecl _mm256_broadcast_ss(const float *);
extern void __cdecl _mm256_store_ps(float *, __m256);
}
#endif
#else
#if defined(TERATHON_SSE)
#include <emmintrin.h>
#endif
#if defined(TERATHON_AVX)
#include <immintrin.h>
#endif
#if defined(TERATHON_NEON)
#include <arm_neon.h>
#endif
#endif
namespace Terathon
{
struct vec_float
{
#if defined(TERATHON_SSE)
__m128 v;
vec_float(__m128 m) {v = m;}
operator __m128&(void) {return (v);}
operator const __m128&(void) const {return (v);}
operator volatile __m128&(void) volatile {return (v);}
operator const volatile __m128&(void) const volatile {return (v);}
#elif defined(TERATHON_NEON)
float32x4_t v;
vec_float(float32x4_t m) {v = m;}
operator float32x4_t&(void) {return (v);}
operator const float32x4_t&(void) const {return (v);}
operator volatile float32x4_t&(void) volatile {return (v);}
operator const volatile float32x4_t&(void) const volatile {return (v);}
#else
float v[4];
#endif
inline vec_float() = default;
inline vec_float(const vec_float&) = default;
inline vec_float& operator =(const vec_float& m) = default;
};
struct exv_float
{
#if defined(TERATHON_AVX)
__m256 v;
exv_float(__m256 m) {v = m;}
operator __m256&(void) {return (v);}
operator const __m256&(void) const {return (v);}
operator volatile __m256&(void) volatile {return (v);}
operator const volatile __m256&(void) const volatile {return (v);}
#else
float v[8];
#endif
inline exv_float() = default;
inline exv_float(const exv_float&) = default;
inline exv_float& operator =(const exv_float&) = default;
};
#if defined(TERATHON_SSE)
typedef __m128i vec_int8;
typedef __m128i vec_int16;
typedef __m128i vec_int32;
typedef __m128i vec_uint8;
typedef __m128i vec_uint16;
typedef __m128i vec_uint32;
#elif defined(TERATHON_NEON)
typedef int8x16_t vec_int8;
typedef int16x8_t vec_int16;
typedef int32x4_t vec_int32;
typedef uint8x16_t vec_uint8;
typedef uint16x8_t vec_uint16;
typedef uint32x4_t vec_uint32;
#endif
#if defined(TERATHON_AVX)
typedef __m256i exv_int8;
typedef __m256i exv_int16;
typedef __m256i exv_int32;
typedef __m256i exv_uint8;
typedef __m256i exv_uint16;
typedef __m256i exv_uint32;
#endif
inline vec_float VecFloatGetZero(void)
{
#if defined(TERATHON_SSE)
return (_mm_setzero_ps());
#elif defined(TERATHON_NEON)
return (vmovq_n_f32(0.0F));
#endif
}
inline vec_float VecFloatGetMinusZero(void)
{
#if defined(TERATHON_SSE)
alignas(16) static const uint32 float_80000000[4] = {0x80000000, 0x80000000, 0x80000000, 0x80000000};
return (_mm_load_ps(reinterpret_cast<const float *>(float_80000000)));
#elif defined(TERATHON_NEON)
return (vmovq_n_f32(-0.0F));
#endif
}
template <uint32 value>
inline vec_float VecLoadVectorConstant(void)
{
#if defined(TERATHON_SSE)
alignas(16) static const uint32 k[4] = {value, value, value, value};
return (_mm_load_ps(reinterpret_cast<const float *>(k)));
#elif defined(TERATHON_NEON)
return (vreinterpretq_f32_u32(vdupq_n_u32(value)));
#endif
}
template <uint32 value>
inline vec_float VecLoadScalarConstant(void)
{
#if defined(TERATHON_SSE)
static const uint32 k = value;
return (_mm_load_ss(reinterpret_cast<const float *>(&k)));
#elif defined(TERATHON_NEON)
return (vreinterpretq_f32_u32(vdupq_n_u32(value)));
#endif
}
inline vec_float VecSmearX(const vec_float& v)
{
#if defined(TERATHON_SSE)
return (_mm_shuffle_ps(v, v, _MM_SHUFFLE(0, 0, 0, 0)));
#elif defined(TERATHON_NEON)
return (vdupq_n_f32(vgetq_lane_f32(v, 0)));
#endif
}
inline vec_float VecSmearY(const vec_float& v)
{
#if defined(TERATHON_SSE)
return (_mm_shuffle_ps(v, v, _MM_SHUFFLE(1, 1, 1, 1)));
#elif defined(TERATHON_NEON)
return (vdupq_n_f32(vgetq_lane_f32(v, 1)));
#endif
}
inline vec_float VecSmearZ(const vec_float& v)
{
#if defined(TERATHON_SSE)
return (_mm_shuffle_ps(v, v, _MM_SHUFFLE(2, 2, 2, 2)));
#elif defined(TERATHON_NEON)
return (vdupq_n_f32(vgetq_lane_f32(v, 2)));
#endif
}
inline vec_float VecSmearW(const vec_float& v)
{
#if defined(TERATHON_SSE)
return (_mm_shuffle_ps(v, v, _MM_SHUFFLE(3, 3, 3, 3)));
#elif defined(TERATHON_NEON)
return (vdupq_n_f32(vgetq_lane_f32(v, 3)));
#endif
}
#if defined(TERATHON_SSE)
template <int p3, int p2, int p1, int p0>
inline vec_float VecShuffle(const vec_float& v1, const vec_float& v2)
{
return (_mm_shuffle_ps(v1, v2, _MM_SHUFFLE(p3, p2, p1, p0)));
}
#endif
inline vec_float VecLoad(const float *ptr)
{
#if defined(TERATHON_SSE)
return (_mm_load_ps(ptr));
#elif defined(TERATHON_NEON)
return (vld1q_f32(ptr));
#endif
}
inline vec_float VecLoadUnaligned(const float *ptr)
{
#if defined(TERATHON_SSE)
return (_mm_loadu_ps(ptr));
#elif defined(TERATHON_NEON)
return (vld1q_f32(ptr));
#endif
}
inline vec_float VecLoadScalar(const float *ptr)
{
#if defined(TERATHON_SSE)
return (_mm_load_ss(ptr));
#elif defined(TERATHON_NEON)
return (vld1q_dup_f32(ptr));
#endif
}
inline vec_float VecLoadSmearScalar(const float *ptr)
{
#if defined(TERATHON_SSE)
vec_float v = _mm_load_ss(ptr);
return (_mm_shuffle_ps(v, v, _MM_SHUFFLE(0, 0, 0, 0)));
#elif defined(TERATHON_NEON)
return (vld1q_dup_f32(ptr));
#endif
}
inline void VecStore(const vec_float& v, float *ptr)
{
#if defined(TERATHON_SSE)
_mm_store_ps(ptr, v);
#elif defined(TERATHON_NEON)
vst1q_f32(ptr, v);
#endif
}
inline void VecStoreUnaligned(const vec_float& v, float *ptr)
{
#if defined(TERATHON_SSE)
_mm_storeu_ps(ptr, v);
#elif defined(TERATHON_NEON)
vst1q_f32(ptr, v);
#endif
}
inline void VecStoreX(const vec_float& v, float *ptr)
{
#if defined(TERATHON_SSE4)
*ptr = v[0];
#elif defined(TERATHON_SSE)
_mm_store_ss(ptr, v);
#elif defined(TERATHON_NEON)
vst1q_lane_f32(ptr, v, 0);
#endif
}
inline void VecStoreY(const vec_float& v, float *ptr)
{
#if defined(TERATHON_SSE4)
*ptr = v[1];
#elif defined(TERATHON_SSE)
_mm_store_ss(ptr, _mm_shuffle_ps(v, v, _MM_SHUFFLE(1, 1, 1, 1)));
#elif defined(TERATHON_NEON)
vst1q_lane_f32(ptr, v, 1);
#endif
}
inline void VecStoreZ(const vec_float& v, float *ptr)
{
#if defined(TERATHON_SSE4)
*ptr = v[2];
#elif defined(TERATHON_SSE)
_mm_store_ss(ptr, _mm_shuffle_ps(v, v, _MM_SHUFFLE(2, 2, 2, 2)));
#elif defined(TERATHON_NEON)
vst1q_lane_f32(ptr, v, 2);
#endif
}
inline void VecStoreW(const vec_float& v, float *ptr)
{
#if defined(TERATHON_SSE4)
*ptr = v[3];
#elif defined(TERATHON_SSE)
_mm_store_ss(ptr, _mm_shuffle_ps(v, v, _MM_SHUFFLE(3, 3, 3, 3)));
#elif defined(TERATHON_NEON)
vst1q_lane_f32(ptr, v, 3);
#endif
}
inline void VecStore3D(const vec_float& v, float *ptr)
{
#if defined(TERATHON_SSE4)
ptr[0] = v[0];
ptr[1] = v[1];
ptr[2] = v[2];
#elif defined(TERATHON_SSE)
_mm_store_ss(&ptr[0], v);
_mm_store_ss(&ptr[1], _mm_shuffle_ps(v, v, _MM_SHUFFLE(1, 1, 1, 1)));
_mm_store_ss(&ptr[2], _mm_shuffle_ps(v, v, _MM_SHUFFLE(2, 2, 2, 2)));
#elif defined(TERATHON_NEON)
vst1q_lane_f32(&ptr[0], v, 0);
vst1q_lane_f32(&ptr[1], v, 1);
vst1q_lane_f32(&ptr[2], v, 2);
#endif
}
inline int32 VecTruncateConvert(const vec_float& v)
{
#if defined(TERATHON_SSE)
return (_mm_cvtt_ss2si(v));
#elif defined(TERATHON_NEON)
return (vgetq_lane_s32(vcvtq_s32_f32(v), 0));
#endif
}
inline vec_float VecNegate(const vec_float& v)
{
#if defined(TERATHON_SSE)
return (_mm_sub_ps(_mm_setzero_ps(), v));
#elif defined(TERATHON_NEON)
return (vnegq_f32(v));
#endif
}
inline vec_float operator -(const vec_float& v)
{
#if defined(TERATHON_SSE)
return (_mm_sub_ps(_mm_setzero_ps(), v));
#elif defined(TERATHON_NEON)
return (vnegq_f32(v));
#endif
}
inline vec_float VecMin(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_min_ps(v1, v2));
#elif defined(TERATHON_NEON)
return (vminq_f32(v1, v2));
#endif
}
inline vec_float VecMinScalar(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_min_ss(v1, v2));
#elif defined(TERATHON_NEON)
return (vminq_f32(v1, v2));
#endif
}
inline vec_float VecMax(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_max_ps(v1, v2));
#elif defined(TERATHON_NEON)
return (vmaxq_f32(v1, v2));
#endif
}
inline vec_float VecMaxScalar(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_max_ss(v1, v2));
#elif defined(TERATHON_NEON)
return (vmaxq_f32(v1, v2));
#endif
}
inline vec_float VecAdd(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_add_ps(v1, v2));
#elif defined(TERATHON_NEON)
return (vaddq_f32(v1, v2));
#endif
}
inline vec_float operator +(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_add_ps(v1, v2));
#elif defined(TERATHON_NEON)
return (vaddq_f32(v1, v2));
#endif
}
inline vec_float VecAddScalar(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_add_ss(v1, v2));
#elif defined(TERATHON_NEON)
return (vaddq_f32(v1, v2));
#endif
}
inline vec_float VecSub(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_sub_ps(v1, v2));
#elif defined(TERATHON_NEON)
return (vsubq_f32(v1, v2));
#endif
}
inline vec_float operator -(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_sub_ps(v1, v2));
#elif defined(TERATHON_NEON)
return (vsubq_f32(v1, v2));
#endif
}
inline vec_float VecSubScalar(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_sub_ss(v1, v2));
#elif defined(TERATHON_NEON)
return (vsubq_f32(v1, v2));
#endif
}
inline vec_float VecMul(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_mul_ps(v1, v2));
#elif defined(TERATHON_NEON)
return (vmulq_f32(v1, v2));
#endif
}
inline vec_float operator *(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_mul_ps(v1, v2));
#elif defined(TERATHON_NEON)
return (vmulq_f32(v1, v2));
#endif
}
inline vec_float VecMulScalar(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_mul_ss(v1, v2));
#elif defined(TERATHON_NEON)
return (vmulq_f32(v1, v2));
#endif
}
inline vec_float VecMadd(const vec_float& v1, const vec_float& v2, const vec_float& v3)
{
#if defined(TERATHON_SSE)
return (_mm_add_ps(_mm_mul_ps(v1, v2), v3));
#elif defined(TERATHON_NEON)
return (vfmaq_f32(v3, v1, v2));
#endif
}
inline vec_float VecMaddScalar(const vec_float& v1, const vec_float& v2, const vec_float& v3)
{
#if defined(TERATHON_SSE)
return (_mm_add_ss(_mm_mul_ss(v1, v2), v3));
#elif defined(TERATHON_NEON)
return (vfmaq_f32(v3, v1, v2));
#endif
}
inline vec_float VecNmsub(const vec_float& v1, const vec_float& v2, const vec_float& v3)
{
#if defined(TERATHON_SSE)
return (_mm_sub_ps(v3, _mm_mul_ps(v1, v2)));
#elif defined(TERATHON_NEON)
return (vfmsq_f32(v3, v1, v2));
#endif
}
inline vec_float VecNmsubScalar(const vec_float& v1, const vec_float& v2, const vec_float& v3)
{
#if defined(TERATHON_SSE)
return (_mm_sub_ss(v3, _mm_mul_ss(v1, v2)));
#elif defined(TERATHON_NEON)
return (vfmsq_f32(v3, v1, v2));
#endif
}
inline vec_float VecDiv(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_div_ps(v1, v2));
#elif defined(TERATHON_NEON)
vec_float f = vrecpeq_f32(v2);
f = vmulq_f32(f, vrecpsq_f32(v2, f));
return (vmulq_f32(v1, f));
#endif
}
inline vec_float operator /(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_div_ps(v1, v2));
#elif defined(TERATHON_NEON)
vec_float f = vrecpeq_f32(v2);
f = vmulq_f32(f, vrecpsq_f32(v2, f));
return (vmulq_f32(v1, f));
#endif
}
inline vec_float VecDivScalar(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_div_ss(v1, v2));
#elif defined(TERATHON_NEON)
vec_float f = vrecpeq_f32(v2);
f = vmulq_f32(f, vrecpsq_f32(v2, f));
return (vmulq_f32(v1, f));
#endif
}
inline vec_float VecAnd(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_and_ps(v1, v2));
#elif defined(TERATHON_NEON)
return (vreinterpretq_f32_u32(vandq_u32(vreinterpretq_u32_f32(v1), vreinterpretq_u32_f32(v2))));
#endif
}
inline vec_float operator &(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_and_ps(v1, v2));
#elif defined(TERATHON_NEON)
return (vreinterpretq_f32_u32(vandq_u32(vreinterpretq_u32_f32(v1), vreinterpretq_u32_f32(v2))));
#endif
}
inline vec_float VecAndc(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_andnot_ps(v2, v1));
#elif defined(TERATHON_NEON)
return (vreinterpretq_f32_u32(vbicq_u32(vreinterpretq_u32_f32(v1), vreinterpretq_u32_f32(v2))));
#endif
}
inline vec_float VecOr(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_or_ps(v1, v2));
#elif defined(TERATHON_NEON)
return (vreinterpretq_f32_u32(vorrq_u32(vreinterpretq_u32_f32(v1), vreinterpretq_u32_f32(v2))));
#endif
}
inline vec_float operator |(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_or_ps(v1, v2));
#elif defined(TERATHON_NEON)
return (vreinterpretq_f32_u32(vorrq_u32(vreinterpretq_u32_f32(v1), vreinterpretq_u32_f32(v2))));
#endif
}
inline vec_float VecXor(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_xor_ps(v1, v2));
#elif defined(TERATHON_NEON)
return (vreinterpretq_f32_u32(veorq_u32(vreinterpretq_u32_f32(v1), vreinterpretq_u32_f32(v2))));
#endif
}
inline vec_float operator ^(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_xor_ps(v1, v2));
#elif defined(TERATHON_NEON)
return (vreinterpretq_f32_u32(veorq_u32(vreinterpretq_u32_f32(v1), vreinterpretq_u32_f32(v2))));
#endif
}
inline vec_float VecSelect(const vec_float& v1, const vec_float& v2, const vec_float& mask)
{
#if defined(TERATHON_SSE)
return (_mm_or_ps(_mm_andnot_ps(mask, v1), _mm_and_ps(mask, v2)));
#elif defined(TERATHON_NEON)
return (vbslq_f32(vreinterpretq_u32_f32(mask), v2, v1));
#endif
}
inline vec_float VecMaskCmpeq(const vec_float& v1, const vec_float& v2)
{
#if defined(TERATHON_SSE)
return (_mm_cmpeq_ps(v1, v2));
#elif defined(TERATHON_NEON)