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
|
/******************************************************************************
Copyright (c) 2007-2024, Intel Corp.
All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:
* Redistributions of source code must retain the above copyright notice,
this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above copyright
notice, this list of conditions and the following disclaimer in the
documentation and/or other materials provided with the distribution.
* Neither the name of Intel Corporation nor the names of its contributors
may be used to endorse or promote products derived from this software
without specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
THE POSSIBILITY OF SUCH DAMAGE.
******************************************************************************/
#if !defined(DPML_UX_H)
#define DPML_UX_H
#if !defined(X_FLOAT)
# define X_FLOAT
#endif
#define NEW_DPML_MACROS 1
#define DPML_PROTOTYPES_H
#include "dpml_private.h"
/* Alignment macros for 16-byte floating point types _Quad and _Decimal128. */
#if defined(_WIN32)||defined(_WIN64)
#define F128_ALIGN_16 __declspec(align(16))
#else
#if !defined(HPUX_OS)
#define F128_ALIGN_16 __attribute__((aligned(16)))
#else
#define F128_ALIGN_16
#endif
#endif
/*
** Types:
**
** Define the basic data types that are used by the unpacked x_float routines
** as well as macros to access their fields and define specific values
*/
typedef INT_32 UX_SIGN_TYPE;
typedef INT_32 UX_EXPONENT_TYPE;
typedef U_INT_32 UX_UNSIGNED_EXPONENT_TYPE;
typedef U_WORD UX_FRACTION_DIGIT_TYPE;
typedef WORD UX_SIGNED_FRACTION_DIGIT_TYPE;
#define BITS_PER_UX_SIGN_TYPE 32
#define BITS_PER_UX_EXPONENT_TYPE 32
#define BITS_PER_UX_FRACTION_DIGIT_TYPE BITS_PER_WORD
#define NUM_UX_FRACTION_DIGITS (128/BITS_PER_UX_FRACTION_DIGIT_TYPE)
#define NUM_X_FRACTION_DIGITS (128/BITS_PER_UX_FRACTION_DIGIT_TYPE)
#if (VAX_FLOATING) || (ENDIANESS == big_endian)
# define DIGIT(n) digit[n]
#else
# define DIGIT(n) digit[NUM_UX_FRACTION_DIGITS - 1 - (n)]
#endif
typedef struct F128_ALIGN_16 {
UX_FRACTION_DIGIT_TYPE digit[ NUM_X_FRACTION_DIGITS ];
} _X_FLOAT;
#define G_X_DIGIT(p,n) (((_X_FLOAT *)(p))->DIGIT(n))
#define P_X_DIGIT(p,n,v) (((_X_FLOAT *)(p))->DIGIT(n) = (v))
#define X_TOGGLE_SIGN(p,v) (((_X_FLOAT *)(p))->DIGIT(0) ^= (v))
#define SHIFT F_EXP_WIDTH
#define CSHIFT (BITS_PER_UX_FRACTION_DIGIT_TYPE - F_EXP_WIDTH)
typedef struct {
UX_SIGN_TYPE sign;
UX_EXPONENT_TYPE exponent;
UX_FRACTION_DIGIT_TYPE fraction[ NUM_UX_FRACTION_DIGITS ];
} UX_FLOAT;
typedef struct {
UX_FRACTION_DIGIT_TYPE digits[ NUM_UX_FRACTION_DIGITS ];
} FIXED_128;
#define UX_SIGN_SHIFT (BITS_PER_UX_FRACTION_DIGIT_TYPE - BITS_PER_UX_SIGN_TYPE)
#define UX_PRECISION 128
#define LSD_NUM (NUM_UX_FRACTION_DIGITS - 1)
#define MSD_NUM 0
#define G_UX_SIGN(x) (((UX_FLOAT*)(x))->sign)
#define G_UX_EXPONENT(x) (((UX_FLOAT*)(x))->exponent)
#define G_UX_MSD(x) (((UX_FLOAT*)(x))->fraction[0])
#define G_UX_2nd_MSD(x) (((UX_FLOAT*)(x))->fraction[1])
#define G_UX_LSD(x) (((UX_FLOAT*)(x))->fraction[LSD_NUM])
#define G_UX_2nd_LSD(x) (((UX_FLOAT*)(x))->fraction[LSD_NUM-1])
#define G_UX_FRACTION_DIGIT(x,n) (((UX_FLOAT*)(x))->fraction[n])
#define P_UX_SIGN(x,v) ((((UX_FLOAT*)(x))->sign)=(v))
#define P_UX_EXPONENT(x,v) ((((UX_FLOAT*)(x))->exponent)=(v))
#define P_UX_MSD(x,v) ((((UX_FLOAT*)(x))->fraction[0])=(v))
#define P_UX_2nd_MSD(x,v) ((((UX_FLOAT*)(x))->fraction[1])=(v))
#define P_UX_LSD(x,v) ((((UX_FLOAT*)(x))->fraction[LSD_NUM])=(v))
#define P_UX_2nd_LSD(x,v) ((((UX_FLOAT*)(x))->fraction[LSD_NUM-1])=(v))
#define P_UX_FRACTION_DIGIT(x,n,v) (((UX_FLOAT*)(x))->fraction[n] = (v))
#define UX_INCR_EXPONENT(x,v) ((((UX_FLOAT *)(x))->exponent) += (v))
#define UX_DECR_EXPONENT(x,v) ((((UX_FLOAT *)(x))->exponent) -= (v))
#define UX_TOGGLE_SIGN(x,v) ((((UX_FLOAT *)(x))->sign) ^= (v))
#define MINUS_ONE 0xFFFFFFFF
#define UX_SIGN_BIT ((U_WORD) 1 << 31)
#define UX_MSB ((U_WORD)1 <<(BITS_PER_UX_FRACTION_DIGIT_TYPE-1))
#define UX_OVERFLOW_EXPONENT (1 << F_EXP_WIDTH)
#define UX_UNDERFLOW_EXPONENT (- UX_OVERFLOW_EXPONENT)
#define UX_ZERO_EXPONENT (MINUS_ONE << (F_EXP_WIDTH + 2))
#define UX_INFINITY_EXPONENT (-(UX_ZERO_EXPONENT + 1))
#define AS_DIGIT(p,n) (((UX_FRACTION_DIGIT_TYPE *)(p))[n])
#include "dpml_ux_32_64.h"
#define UX_LOW_FRACTION_IS_ZERO(p) (UX_OR_LOW_FRACTION_DIGITS(p) == 0)
#define UX_FRACTION_IS_ONE_HALF(p) ((G_UX_MSD(p) == UX_MSB) & \
(UX_OR_LOW_FRACTION_DIGITS(p) == 0))
#define UX_SET_SIGN_EXP_MSD(p,s,e,m) ( P_UX_SIGN(p,s), \
P_UX_EXPONENT(p,e), \
P_UX_MSD(p,m), \
CLR_UX_LOW_FRACTION(p))
#define UX_COPY(p,q) ( P_UX_SIGN(q, G_UX_SIGN(p)), \
P_UX_EXPONENT(q, G_UX_EXPONENT(p)), \
COPY_TO_UX_FRACTION(&G_UX_MSD(p),q))
typedef U_WORD ERROR_CODE;
/******************************************************************************/
/******************************************************************************/
/** **/
/** Name Macros **/
/** **/
/******************************************************************************/
/******************************************************************************/
/*
** Following macros are defined to modify the interface of X_FLOAT routines
** for different architectures variants at compile time. The macros are
** defined as returnType_Arg1Arg2_PROTO. For example X_X_PROTO defines a
** function which takes X_FLOAT argument and result is X_FLOAT argument.
**
** X_FLOAT_RES_OR_VOID defines what functions is returning. It can be void,
** X_FLOAT or X_FLOAT *.
**
** X_FLOAT_RET_TYPE(x) defines the return type when result is part of the
** argument list i.e a pointer is provided in the argument list to
** put the result. It can be Nothing or X_FLOAT *x.
**
** X_FLOAT_ARG_TYPE(x) defines the argument type. It can be X_FLOAT *x, or
** X_FLOAT x.
**
** X_FLOAT_INT_TYPE defines the integer type in the argument list. This should
** be int in case of intel compilers.
**
** RETURN_X_FLOAT(x) defines the return statement of the function. It can be
** be Nothing, return *x or return x
*/
#if defined(EMT64_LINUX_QUAD_INTERFACE)
# define X_FLOAT_RET_TYPE(x)
# define X_FLOAT_ARG_TYPE(x) _Quad x
# define X_FLOAT_INT_TYPE int
# define X_FLOAT_RES_OR_VOID _Quad
# define DECLARE_X_FLOAT(res) _X_FLOAT res;
# define PASS_RET_X_FLOAT(x) &x
//# define PASS_ARG_X_FLOAT(x) &x
# define PASS_ARG_X_FLOAT(x) (_X_FLOAT *) &x
# define RETURN_X_FLOAT(x) return *(_Quad *) &(x)
# define PACKED_ARG_IS_NEG(p) ((WORD)((_X_FLOAT *)(&p))->DIGIT(0) < 0)
#elif defined(X_NONVOID_RES_VAL_ARG_VAL)
# define X_FLOAT_RET_TYPE(x)
# define X_FLOAT_ARG_TYPE(x) _X_FLOAT x
# define X_FLOAT_INT_TYPE int
# define X_FLOAT_RES_OR_VOID _X_FLOAT
# define DECLARE_X_FLOAT(res) _X_FLOAT res;
# define PASS_RET_X_FLOAT(x) &x
# define PASS_ARG_X_FLOAT(x) &x
# define RETURN_X_FLOAT(x) return x;
# define PACKED_ARG_IS_NEG(p) ((WORD)((_X_FLOAT *)(&p))->DIGIT(0) < 0)
#elif defined(X_VOID_RES_REF_ARG_VAL)
# define X_FLOAT_RET_TYPE(x) _X_FLOAT *x,
# define X_FLOAT_ARG_TYPE(x) _X_FLOAT x
# define X_FLOAT_INT_TYPE int
# define X_FLOAT_RES_OR_VOID void
# define DECLARE_X_FLOAT(res)
# define PASS_RET_X_FLOAT(x) x
# define PASS_ARG_X_FLOAT(x) &x
# define RETURN_X_FLOAT(x) return;
# define PACKED_ARG_IS_NEG(p) ((WORD)((_X_FLOAT *)(&p))->DIGIT(0) < 0)
#else
# define X_FLOAT_RET_TYPE(x) _X_FLOAT *x,
# define X_FLOAT_ARG_TYPE(x) _X_FLOAT *x
# define X_FLOAT_INT_TYPE WORD
# define X_FLOAT_RES_OR_VOID void
# define DECLARE_X_FLOAT(res)
# define PASS_RET_X_FLOAT(x) x
# define PASS_ARG_X_FLOAT(x) x
# define RETURN_X_FLOAT(x) return;
# define PACKED_ARG_IS_NEG(p) ((WORD)((_X_FLOAT *)(p))->DIGIT(0) < 0)
#endif
#if !defined(X_FLOAT_INT_TYPE)
# define X_FLOAT_INT_TYPE WORD
#endif
# define X_I_PROTO(name,res,arg) \
X_FLOAT_RES_OR_VOID name(X_FLOAT_RET_TYPE(res) int arg)
# define X_X_PROTO(name,res,arg) \
X_FLOAT_RES_OR_VOID name(X_FLOAT_RET_TYPE(res) \
X_FLOAT_ARG_TYPE(arg)) \
# define X_XX_PROTO(name, res, arg1, arg2) \
X_FLOAT_RES_OR_VOID name(X_FLOAT_RET_TYPE(res) \
X_FLOAT_ARG_TYPE(arg1), X_FLOAT_ARG_TYPE(arg2)) \
# define X_XI_PROTO(name, res, arg,i) \
X_FLOAT_RES_OR_VOID name(X_FLOAT_RET_TYPE(res) \
X_FLOAT_ARG_TYPE(arg), X_FLOAT_INT_TYPE i) \
# define X_IX_PROTO(name, res, i, arg) \
X_FLOAT_RES_OR_VOID name(X_FLOAT_RET_TYPE(res) \
X_FLOAT_INT_TYPE i, X_FLOAT_ARG_TYPE(arg)) \
# define X_XIptr_PROTO(name, res, arg,i) \
X_FLOAT_RES_OR_VOID name(X_FLOAT_RET_TYPE(res) \
X_FLOAT_ARG_TYPE(arg), X_FLOAT_INT_TYPE *i) \
# define X_XXptr_PROTO(name, res, arg, p) \
X_FLOAT_RES_OR_VOID name(X_FLOAT_RET_TYPE(res) \
X_FLOAT_ARG_TYPE(arg), _X_FLOAT * p) \
# define X_XXIptr_PROTO(name, res, arg1, arg2, i) \
X_FLOAT_RES_OR_VOID name(X_FLOAT_RET_TYPE(res) \
X_FLOAT_ARG_TYPE(arg1), X_FLOAT_ARG_TYPE(arg2), \
X_FLOAT_INT_TYPE *i) \
# define I_XXI_PROTO(name, arg1, arg2, i) \
int name( X_FLOAT_ARG_TYPE(arg1), X_FLOAT_ARG_TYPE(arg2), int i)
# define RR_X_PROTO(name, res1, res2, arg) \
void name(X_FLOAT_ARG_TYPE(arg), _X_FLOAT *res1, _X_FLOAT *res2)
/******************************************************************************/
/******************************************************************************/
/** **/
/** Packed and Unpacked Constant Tables **/
/** **/
/******************************************************************************/
/******************************************************************************/
#if !defined(PACKED_CONSTANT_TABLE)
# define PACKED_CONSTANT_TABLE __TABLE_NAME(x_constants__ )
#endif
#undef NAN
#if !defined(DPML_UX_CONS_FILE_NAME)
# define DPML_UX_CONS_FILE_NAME dpml_cons_x.h
#endif
#if !defined(BUILD_UX_CONS_TABLE)
# define INSTANTIATE_TABLE 0
# define INSTANTIATE_DEFINES 1
# include STR(DPML_UX_CONS_FILE_NAME)
#endif
/******************************************************************************/
/******************************************************************************/
/** **/
/** Pack and Unpacked Routines **/
/** **/
/******************************************************************************/
/******************************************************************************/
/*
** There's a slight complication here because the interface to these routines
** depend on the interface to the exception handler. Specifically, if we
** need to pass the original arguments and/or the name of the function to the
** exception handler, that information must be passed to pack and unpack.
*/
typedef struct {
U_WORD arg_classes;
char * name;
_X_FLOAT * args[2];
} UX_EXCEPTION_INFO_STRUCT;
#define EXCPTN_INFO __INTERNAL_NAME(ux_excptn_info__)
#define EXCEPTION_INFO_DECL UX_EXCEPTION_INFO_STRUCT EXCEPTION_INFO;
#define OPT_EXCEPTION_INFO , &EXCEPTION_INFO
#define OPT_EXCEPTION_INFO_DECLARATION , UX_EXCEPTION_INFO_STRUCT * EXCPTN_INFO
#define OPT_EXCEPTION_INFO_ARGUMENT , EXCPTN_INFO
#define IF_OPTNL_ERROR_INFO(x) x
#if (EXCEPTION_INTERFACE_SEND & send_function_name )
# define INIT_EXCEPTION_INFO EXCPTN_INFO.name = STR(F_ENTRY_NAME)
#else
# define INIT_EXCEPTION_INFO
#endif
#define UNPACK(a,b,c,d) UNPACK_X_OR_Y(a,0,b,c,d)
#if !defined( UNPACK_X_OR_Y )
# define UNPACK_X_OR_Y __INTERNAL_NAME( unpack_x_or_y__ )
#endif
#if !defined( UNPACK2 )
# define UNPACK2 __INTERNAL_NAME( unpack2__ )
#endif
#if !defined( PACK )
# define PACK __INTERNAL_NAME( pack__ )
#endif
extern WORD UNPACK_X_OR_Y (
_X_FLOAT *, /* packed argument 1 */
_X_FLOAT *, /* packed argument 2 */
UX_FLOAT *, /* unpacked argument */
U_WORD const *, /* class-to-action map */
_X_FLOAT * /* packed result */
OPT_EXCEPTION_INFO_DECLARATION
);
extern WORD UNPACK2 (
_X_FLOAT *, /* packed argument 1 */
_X_FLOAT *, /* packed argument 2 */
UX_FLOAT *, /* unpacked argument 1 */
UX_FLOAT *, /* unpacked argument 2 */
U_WORD const *, /* class-to-action map */
_X_FLOAT * /* packed result */
OPT_EXCEPTION_INFO_DECLARATION
);
extern void PACK (
UX_FLOAT *, /* unpacked result */
_X_FLOAT *, /* packed result */
ERROR_CODE, /* underflow code */
ERROR_CODE /* overflow code */
OPT_EXCEPTION_INFO_DECLARATION
);
/*
** Include the class-to-action-mapping definitions here, since they are used
** primarily by the unpack routines.
*/
#define INDEX_POS 0
#define INDEX_WIDTH 3
#define INDEX_MASK 0x7
#define ACTION_POS 3
#define ACTION_WIDTH 3
#define ACTION_MASK 0x7
#define CLASS_TO_ACTION(class, action, index) \
(((action << INDEX_WIDTH) | (index)) << \
((INDEX_WIDTH + ACTION_WIDTH)*(class)))
#define CLASS_TO_ACTION_DISP(n) \
((n) << ((INDEX_WIDTH + ACTION_WIDTH)*F_C_NUM_CLASSES))
#define RETURN_UNPACKED 0
#define RETURN_QUIET_NAN 1
#define RETURN_VALUE 2
#define RETURN_NEGATIVE 3
#define RETURN_ABSOLUTE 4
#define RETURN_CPYSN_ARG_0 5
#define RETURN_ERROR 7
#define CLASS_TO_INDEX_WIDTH 4
#define CLASS_TO_INDEX(n,m) ((m) << ((n)*CLASS_TO_INDEX_WIDTH))
#define CLASS_TO_INDEX_MASK MAKE_MASK(CLASS_TO_INDEX_WIDTH, 0)
#define WORDS_PER_CLASS_TO_ACTION_MAP (64/BITS_PER_WORD)
/******************************************************************************/
/******************************************************************************/
/** **/
/** Rational Evaluation Routine **/
/** **/
/******************************************************************************/
/******************************************************************************/
#if !defined( EVALUATE_RATIONAL )
# define EVALUATE_RATIONAL __INTERNAL_NAME( evaluate_rational__ )
#endif
extern void EVALUATE_RATIONAL(
UX_FLOAT *, /* Argument */
FIXED_128 *, /* Coefficient array */
U_WORD, /* Number of coefficients */
U_WORD, /* Evaluation flags */
UX_FLOAT * /* Result */
);
#define STANDARD 0x001
#define POST_MULTIPLY 0x002
#define SQUARE_TERM 0x004
#define ALTERNATE_SIGN 0x008
#define NUM_DEN_FIELD_WIDTH 4
#define NO_DIVIDE ((WORD) 1 << (2*NUM_DEN_FIELD_WIDTH))
#define SWAP ((WORD) 2 << (2*NUM_DEN_FIELD_WIDTH))
#define SKIP ((WORD) 4 << (2*NUM_DEN_FIELD_WIDTH))
#define SCALE_WIDTH 6
#define SCALE_POS (BITS_PER_WORD - SCALE_WIDTH)
#define P_SCALE(n) (((WORD) (n)) << SCALE_POS)
#define G_SCALE(n) (((WORD) (n)) >> SCALE_POS)
#define POLY_SHIFT(u,n) ((((UX_FLOAT *)(u))->exponent)*(n))
#define NUMERATOR_FLAGS(n) (n)
#define DENOMINATOR_FLAGS(n) ((n) << NUM_DEN_FIELD_WIDTH)
#if !defined(EVALUATE_PACKED_POLY)
# define EVALUATE_PACKED_POLY __INTERNAL_NAME(evaluate_packed_poly__)
#endif
void
EVALUATE_PACKED_POLY( UX_FLOAT * argument, WORD degree, FIXED_128 * coefs,
U_WORD mask, WORD bias, UX_FLOAT * result);
/******************************************************************************/
/******************************************************************************/
/** **/
/** Rational Evaluation Routine **/
/** **/
/******************************************************************************/
/******************************************************************************/
#if !defined( ADDSUB )
# define ADDSUB __INTERNAL_NAME( addsub__ )
#endif
extern void ADDSUB(
UX_FLOAT *, /* arg1 */
UX_FLOAT *, /* arg2 */
U_WORD, /* operation flags */
UX_FLOAT * /* result */
);
/*
** The logic of the add/sub routine depends on theses symbols have
** these *SPECIFIC* values. !!! DO NOT CHANGE THEM !!!
*/
#define ADD 0
#define SUB 1
#define ADD_SUB 2
#define SUB_ADD 3
#define MAGNITUDE_ONLY 4
#define NO_NORMALIZATION 8
/******************************************************************************/
/******************************************************************************/
/** **/
/** Round to Integer Routine **/
/** **/
/******************************************************************************/
/******************************************************************************/
#if !defined(UX_RND_TO_INT)
# define UX_RND_TO_INT __INTERNAL_NAME(ux_rnd_to_int__)
#endif
extern WORD UX_RND_TO_INT( /* return val is integer part as int */
UX_FLOAT *, /* argument */
WORD, /* rounding mode bit vector */
UX_FLOAT *, /* Integer part as float, ignored if 0 */
UX_FLOAT *); /* fraction part, ignored if 0 */
#define RZ_BIT_VECTOR 0x0000 /* 0000 0000 0000 0000 */
#define RP_BIT_VECTOR 0x00fa /* 0000 0000 1111 1010 */
#define RM_BIT_VECTOR 0xfa00 /* 1111 1010 0000 0000 */
#define RN_BIT_VECTOR 0xa8a8 /* 1010 1000 1010 1000 */
#define RV_BIT_VECTOR 0xaaaa /* 1010 1010 1010 1010 */
#define INTEGER_RESULT 0x10000
#define FRACTION_RESULT 0x20000
/******************************************************************************/
/******************************************************************************/
/** **/
/** Normalization Routines **/
/** **/
/******************************************************************************/
/******************************************************************************/
#if !defined(FFS_AND_SHIFT)
# define FFS_AND_SHIFT __INTERNAL_NAME(ffs_and_shift__)
#endif
extern WORD FFS_AND_SHIFT( /* returns shift count */
UX_FLOAT *, /* source and destination */
U_WORD); /* 'opcode' */
#define FFS_NORMALIZE 0
#define FFS_CVT_WORD 1
#define FFS_CVT_U_WORD 2
#define NORMALIZE(x) FFS_AND_SHIFT(x, FFS_NORMALIZE)
#define WORD_TO_UX(n,x) (P_UX_MSD(x, n), FFS_AND_SHIFT(x, FFS_CVT_WORD))
#define U_WORD_TO_UX(n,x) (P_UX_MSD(x, n), FFS_AND_SHIFT(x, FFS_CVT_U_WORD))
/******************************************************************************/
/******************************************************************************/
/** **/
/** Radian Trig Prototypes **/
/** **/
/******************************************************************************/
/******************************************************************************/
#if !defined(UX_SINCOS)
# define UX_SINCOS __INTERNAL_NAME(ux_sincos)
#endif
extern WORD UX_SINCOS(
UX_FLOAT *, /* unpacked_argument */
WORD, /* octant */
WORD, /* function_code, */
UX_FLOAT *); /* unpacked_result */
#define DEGREE 16
#define SIN_FUNC 1
#define COS_FUNC 2
#define SINCOS_FUNC (SIN_FUNC | COS_FUNC)
#define SIND_FUNC (SIN_FUNC | DEGREE)
#define COSD_FUNC (COS_FUNC | DEGREE)
#define SINCOSD_FUNC (SINCOS_FUNC | DEGREE)
#define TAN_FUNC 4
#define COT_FUNC 8
#define TAND_FUNC (TAN_FUNC | DEGREE)
#define COTD_FUNC (COT_FUNC | DEGREE)
#define SIN(a,b) EVAL_SINCOS(a, 0, SIN_FUNC, b)
#define COS(a,b) EVAL_SINCOS(a, 0, COS_FUNC, b)
#define SINCOS(a,b) EVAL_SINCOS(a, 0, SINCOS_FUNC, b)
#define SINCOS_COEF_ARRAY_LENGTH 12
extern FIXED_128 sincos_coef_array[2*SINCOS_COEF_ARRAY_LENGTH];
/******************************************************************************/
/******************************************************************************/
/** **/
/** Log Prototypes **/
/** **/
/******************************************************************************/
/******************************************************************************/
#if !defined(UX_LOG)
# define UX_LOG __INTERNAL_NAME( ux_log__ )
#endif
extern void UX_LOG(
UX_FLOAT *, /* Argument */
UX_FLOAT *, /* scale - LOG(x) = scale*log2(x) */
UX_FLOAT *); /* Result */
#define LOG(a,b) UX_LOG( a, & UX_CON( LN_2 ), b)
#if !defined(UX_LOG_POLY)
# define UX_LOG_POLY __INTERNAL_NAME( ux_log_poly__ )
#endif
extern void UX_LOG_POLY(
UX_FLOAT *, /* Argument */
UX_FLOAT *); /* Result */
/******************************************************************************/
/******************************************************************************/
/** **/
/** Miscellaneous Prototypes **/
/** **/
/******************************************************************************/
/******************************************************************************/
#if !defined( EXP )
# define UX_EXP __INTERNAL_NAME( ux_exp__ )
#endif
extern void UX_EXP(
UX_FLOAT *, /* argument */
UX_FLOAT * /* result */
);
#if !defined( DIVIDE )
# define DIVIDE __INTERNAL_NAME( divide__ )
#endif
extern void DIVIDE(
UX_FLOAT *, /* numerator - assume 1 if ptr is 0 */
UX_FLOAT *, /* denominator */
U_WORD, /* result precision */
UX_FLOAT * /* result */
);
#define HALF_PRECISION 1
#define FULL_PRECISION 2
#if !defined( MULTIPLY )
# define MULTIPLY __INTERNAL_NAME( multiply__ )
#endif
extern void MULTIPLY(
UX_FLOAT *, /* arg1 */
UX_FLOAT *, /* arg1 */
UX_FLOAT * /* result */
);
#define SQUARE(a,b) MULTIPLY(a, a, b)
#if !defined( EXTENDED_MULTIPLY )
# define EXTENDED_MULTIPLY __INTERNAL_NAME( extended_multiply__ )
#endif
extern void EXTENDED_MULTIPLY(
UX_FLOAT *, /* arg1 */
UX_FLOAT *, /* arg1 */
UX_FLOAT *, /* hi result */
UX_FLOAT * /* lo result */
);
#if !defined(UX_SQRT_EVALUATION)
# define UX_SQRT_EVALUATION __INTERNAL_NAME( ux_sqrt_evaluation__ )
#endif
#define EVALUATE_SQRT 0
#define EVALUATE_RSQRT 1
extern void UX_SQRT_EVALUATION(
UX_FLOAT *, /* Argument */
WORD, /* evaluation type - sqrt or rsqrt */
UX_FLOAT *); /* Result */
#define UX_SQRT(a,b) UX_SQRT_EVALUATION(a, EVALUATE_SQRT, b)
#if !defined(HYPOT)
# define HYPOT __INTERNAL_NAME( hypot__ )
#endif
extern void HYPOT(
UX_FLOAT *, /* Argument 1 */
UX_FLOAT *, /* Argument 2 */
UX_FLOAT *); /* Result */
/******************************************************************************/
/******************************************************************************/
/** **/
/** Miscellaneous Definitions **/
/** **/
/******************************************************************************/
/******************************************************************************/
#define NONE 0
#define NOT_USED 0
#if defined(NULL)
# undef NULL
#endif
#define NULL 0
#if defined GROUP
# define D_GROUP(x) GROUP(x)
#else
# define D_GROUP_NAME PASTE_2(__INTERNAL_NAME(group),_d)
extern double D_GROUP_NAME( double );
# define D_GROUP(x) D_GROUP_NAME(x)
#endif
/******************************************************************************/
/******************************************************************************/
/** **/
/** MPHOC Macros for Class-to-Action Table Definitions **/
/** **/
/******************************************************************************/
/******************************************************************************/
#define POS 0
#define NEG UX_SIGN_BIT
#if defined(MAKE_INCLUDE)
# define PRINT_64_TBL_ITEM(i) \
printf( "\t/* %3i */ %#16.4.16i,\n", BYTES(MP_BIT_OFFSET), i);\
MP_BIT_OFFSET += 64
# define PRINT_UX_FRACTION_DIGIT_TBL_ITEM(val) PRINT_64_TBL_ITEM(val)
# define PRINT_CLASS_TO_ACTION_TBL_DEF(name) \
printf("#define\t" name "\t((U_WORD const *) ((char *) "\
STR(MP_TABLE_NAME) " + %i))\n", BYTES(MP_BIT_OFFSET))
# define PRINT_UX_FRACTION_DIGIT_TBL_VDEF(name) \
printf("#define\t" name \
"\t*((UX_FRACTION_DIGIT_TYPE *) ((char *) " \
STR(MP_TABLE_NAME) " + %i))\n", BYTES(MP_BIT_OFFSET))
# define PRINT_UX_FRACTION_DIGIT_TBL_VDEF_ITEM(name, val) \
printf("#define\t" name \
"\t*((UX_FRACTION_DIGIT_TYPE *) ((char *) " \
STR(MP_TABLE_NAME) " + %i))\n", BYTES(MP_BIT_OFFSET)); \
PRINT_64_TBL_ITEM(val)
# define PRINT_FIXED_128_TBL_ADEF(name) \
printf("#define\t" name "\t((FIXED_128 *) ((char *) " \
STR(MP_TABLE_NAME) " + %i))\n", BYTES(MP_BIT_OFFSET))
# define PRINT_UX_FRACTION_DIGIT_TBL_ADEF(name) \
printf("#define\t" name "\t((UX_FRACTION_DIGIT_TYPE *) ((char *) " \
STR(MP_TABLE_NAME) " + %i))\n", BYTES(MP_BIT_OFFSET))
# define PRINT_UX_TBL_ADEF(name) \
printf("#define\t" name "\t((UX_FLOAT *) ((char *) " \
STR(MP_TABLE_NAME) " + %i))\n", BYTES(MP_BIT_OFFSET))
# define PRINT_UX_TBL_ITEM(val) \
MP_BIT_OFFSET = print_ux_table_value(val, MP_BIT_OFFSET)
# define PRINT_UX_TBL_ADEF_ITEM(name, val) \
PRINT_UX_TBL_ADEF(name); PRINT_UX_TBL_ITEM(val)
@divert divertText
function print_ux_fraction_digits(value)
{
auto hi, i;
if (value >= 1)
{
printf("ERROR: value out of range in print_ux_fraction_digits\n");
exit;
}
for (i = NUM_UX_FRACTION_DIGITS; i > 0; i--)
{
value = bldexp(value, BITS_PER_UX_FRACTION_DIGIT_TYPE);
hi = trunc(value);
if (hi)
printf( DIGIT_FORMAT, hi);
else
printf( ZERO_FORMAT );
value -= hi;
}
printf("\n");
return value;
}
function print_ux_table_value(value, offset)
{
auto exponent, hi, sign_bit, i;
sign_bit = 0;
if (value == 0)
exponent = bldexp(-1, F_EXP_WIDTH + 2);
else
{
exponent = bexp(value);
if (value < 0)
{
sign_bit = 1;
value = -value;
}
value = bldexp(value, -exponent);
}
if (sign_bit)
printf("\t/* %3i */ NEG, %4i,", BYTES(offset), exponent);
else
printf("\t/* %3i */ POS, %4i,", BYTES(offset), exponent);
print_ux_fraction_digits(value);
return offset + BITS_PER_UX_SIGN_TYPE + BITS_PER_UX_EXPONENT_TYPE +
UX_PRECISION;
}
function find_max_exponent(degree, index)
{
auto i, max_exponent;
max_exponent = -128;
for (i = 0; i <= degree; i++)
{
exponent = bexp(bround(ux_rational_coefs[index + i], 128));
if (exponent > max_exponent)
max_exponent = exponent;
}
return max_exponent;
}
/*
** The following routine prints out coefficients in the array
** 'ux_rational_coefs' in fixed point format.
*/
procedure print_ux_poly_coefs(pad_len, degree, final_scale, index)
{
auto scale, exponent, i;
for (i = 0; i < pad_len; i++)
{
printf( "\t/* %3i */ %#32.4.16i,\n", BYTES(MP_BIT_OFFSET), 0);
MP_BIT_OFFSET += 128;
}
exponent = find_max_exponent(degree, index);
scale = 128 - exponent;
index += degree;
for (i = degree; i >= 0; i--)
{
printf( "\t/* %3i */ %#32.4.16i,\n", BYTES(MP_BIT_OFFSET),
abs(nint(bldexp(ux_rational_coefs[index], scale ))));
MP_BIT_OFFSET += 128;
index--;
}
PRINT_U_TBL_ITEM(exponent + final_scale);
}
function print_ux_rational_coefs( num_degree, den_degree, scale)
{
auto max_degree;
max_degree = max(num_degree, den_degree);
print_ux_poly_coefs(max_degree - num_degree , num_degree, scale, 0);
if (den_degree)
print_ux_poly_coefs(max_degree - den_degree, den_degree, 0,
num_degree + 1);
return max_degree;
}
/*
** This routine finds the "width" and "bias" for converting MP numbers
** to a special 128 bit packed format used for special polynomial
** evaluations. The coefficients are contained in the global array
** ux_rational_coef and the both the width and the bias are returned
** via global values. See the description in dpml_ux_ops.c
*/
procedure find_exponent_width_and_bias(degree, base_index)
{
auto i, top, _diff, min_diff, max_diff, old_exp, new_exp, width;
top = base_index + degree;
min_diff = max_diff = 0;
old_exp = 0;
for (i = base_index; i <= top; i++)
{
new_exp = bexp(ux_rational_coefs[i]);
_diff = new_exp - old_exp;
if (_diff < min_diff)
min_diff = _diff;
else if (_diff > max_diff)
max_diff = _diff;
old_exp = new_exp;
}
_diff = max_diff - min_diff + 1;
width = bexp(_diff);
if (bldexp(.5, width) == _diff)
width--;
packed_exponent_width = width;
packed_exponent_bias = -min_diff;
}
/*
** After we know the bias and the width, we need to pack the coefficient
** values
*/
procedure cvt_to_packed(degree, base_index, width, bias)
{
auto i, top, num_bits, tmp, old_exp, new_exp, sign_bit;
find_exponent_width_and_bias(degree, base_index);
top = base_index + degree;
old_exp = 0;
num_bits = UX_PRECISION - width - 1;
for (i = base_index; i <= top; i++)
{
sign_bit = 0;
tmp = bround(ux_rational_coefs[i], num_bits);
if (tmp < 0)
{
tmp = -tmp;
sign_bit = 1;
}
new_exp = bexp(tmp);
tmp = nint(bldexp(tmp, num_bits - new_exp));
tmp = bldexp(tmp, width + 1) + 2*(new_exp - old_exp + bias) +
sign_bit;
old_exp = new_exp;
ux_rational_coefs[i] = tmp;
}
}
/*
** After converting to pack format, we need to print them out
*/
procedure print_packed(degree, base_index)
{
auto i, top;
top = base_index + degree;
for (i = degree; i >= 0; i--)
{
printf( "\t/* %3i */ %#32.4.16i,\n", BYTES(MP_BIT_OFFSET),
ux_rational_coefs[top--]);
MP_BIT_OFFSET += 128;
}
}
@end_divert
#endif
#if !defined(EXTENDED_DIGIT_MULTIPLY)
# define EXTENDED_DIGIT_MULTIPLY(a,b,h,l) (l) = (a)*(b); UMULH(a,b,h)
#endif
#endif
|
