about summary refs log tree commit diff
path: root/sysdeps/ia64/fpu/s_tan.S
blob: 3000f5ee0670b9bc6607fd0204cafffacbf442bc (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
.file "tancot.s"


// Copyright (c) 2000 - 2003, Intel Corporation
// All rights reserved.
//
// Contributed 2000 by the Intel Numerics Group, Intel Corporation
//
// 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.
//
// * The name of Intel Corporation may not 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 INTEL OR ITS 
// 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. 
//
// Intel Corporation is the author of this code, and requests that all
// problem reports or change requests be submitted to it directly at 
// http://www.intel.com/software/products/opensource/libraries/num.htm.
//
// History
//==============================================================
// 02/02/00 Initial version
// 04/04/00 Unwind support added
// 12/27/00 Improved speed
// 02/21/01 Updated to call tanl
// 05/30/02 Added cot
// 02/10/03 Reordered header: .section, .global, .proc, .align
//
// API
//==============================================================
// double tan(double x);
// double cot(double x);
//
// Overview of operation
//==============================================================
// If the input value in radians is |x| >= 1.xxxxx 2^10 call the
// older slower version.
//
// The new algorithm is used when |x| <= 1.xxxxx 2^9.
//
// Represent the input X as Nfloat * pi/2 + r
//    where r can be negative and |r| <= pi/4
//
//     tan_W  = x * 2/pi
//     Nfloat = round_int(tan_W)
//
//     tan_r  = x - Nfloat * (pi/2)_hi
//  a) tan_r  = tan_r - Nfloat * (pi/2)_lo (for tan)
//  b) tan_r  = Nfloat * (pi/2)_lo - tan_r (for cot)
//
// We have two paths: p8, when Nfloat is even and p9. when Nfloat is odd.
//  a) for tan:  p8: tan(X) =  tan(r)
//               p9: tan(X) = -cot(r)
//  b) for cot:  p9: cot(X) =  cot(r)
//               p8: cot(X) = -tan(r)
//
// Each is evaluated as a series. The p9 path requires 1/r.
//
// The coefficients used in the series are stored in a table as
// are the pi constants.
//
// Registers used
//==============================================================
//
// predicate registers used:
// p6-12
//
// floating-point registers used:
// f10-15, f32-106
// f8, input
//
// general registers used
// r14-26, r32-39
//
// Assembly macros
//==============================================================
TAN_INV_PI_BY_2_2TO64        = f10
TAN_RSHF_2TO64               = f11
TAN_2TOM64                   = f12
TAN_RSHF                     = f13
TAN_W_2TO64_RSH              = f14
TAN_NFLOAT                   = f15

tan_Inv_Pi_by_2              = f32
tan_Pi_by_2_hi               = f33
tan_Pi_by_2_lo               = f34


tan_P0                       = f35
tan_P1                       = f36
tan_P2                       = f37
tan_P3                       = f38
tan_P4                       = f39
tan_P5                       = f40
tan_P6                       = f41
tan_P7                       = f42
tan_P8                       = f43
tan_P9                       = f44
tan_P10                      = f45
tan_P11                      = f46
tan_P12                      = f47
tan_P13                      = f48
tan_P14                      = f49
tan_P15                      = f50

tan_Q0                       = f51
tan_Q1                       = f52
tan_Q2                       = f53
tan_Q3                       = f54
tan_Q4                       = f55
tan_Q5                       = f56
tan_Q6                       = f57
tan_Q7                       = f58
tan_Q8                       = f59
tan_Q9                       = f60
tan_Q10                      = f61

tan_r                        = f62
tan_rsq                      = f63
tan_rcube                    = f64

tan_v18                      = f65
tan_v16                      = f66
tan_v17                      = f67
tan_v12                      = f68
tan_v13                      = f69
tan_v7                       = f70
tan_v8                       = f71
tan_v4                       = f72
tan_v5                       = f73
tan_v15                      = f74
tan_v11                      = f75
tan_v14                      = f76
tan_v3                       = f77
tan_v6                       = f78
tan_v10                      = f79
tan_v2                       = f80
tan_v9                       = f81
tan_v1                       = f82
tan_int_Nfloat               = f83
tan_Nfloat                   = f84

tan_NORM_f8                  = f85
tan_W                        = f86

tan_y0                       = f87
tan_d                        = f88
tan_y1                       = f89
tan_dsq                      = f90
tan_y2                       = f91
tan_d4                       = f92
tan_inv_r                    = f93

tan_z1                       = f94
tan_z2                       = f95
tan_z3                       = f96
tan_z4                       = f97
tan_z5                       = f98
tan_z6                       = f99
tan_z7                       = f100
tan_z8                       = f101
tan_z9                       = f102
tan_z10                      = f103
tan_z11                      = f104
tan_z12                      = f105

arg_copy                     = f106

/////////////////////////////////////////////////////////////

tan_GR_sig_inv_pi_by_2       = r14
tan_GR_rshf_2to64            = r15
tan_GR_exp_2tom64            = r16
tan_GR_n                     = r17
tan_GR_rshf                  = r18
tan_AD                       = r19
tan_GR_10009                 = r20
tan_GR_17_ones               = r21
tan_GR_N_odd_even            = r22
tan_GR_N                     = r23
tan_signexp                  = r24
tan_exp                      = r25
tan_ADQ                      = r26

GR_SAVE_B0                   = r33
GR_SAVE_PFS                  = r34
GR_SAVE_GP                   = r35
GR_Parameter_X               = r36
GR_Parameter_Y               = r37
GR_Parameter_RESULT          = r38
GR_Parameter_Tag             = r39


RODATA

.align 16

LOCAL_OBJECT_START(double_tan_constants)
   data8 0xC90FDAA22168C234, 0x00003FFF // pi/2 hi
   data8 0xBEEA54580DDEA0E1 // P14
   data8 0x3ED3021ACE749A59 // P15
   data8 0xBEF312BD91DC8DA1 // P12
   data8 0x3EFAE9AFC14C5119 // P13
   data8 0x3F2F342BF411E769 // P8
   data8 0x3F1A60FC9F3B0227 // P9
   data8 0x3EFF246E78E5E45B // P10
   data8 0x3F01D9D2E782875C // P11
   data8 0x3F8226E34C4499B6 // P4
   data8 0x3F6D6D3F12C236AC // P5
   data8 0x3F57DA1146DCFD8B // P6
   data8 0x3F43576410FE3D75 // P7
   data8 0x3FD5555555555555 // P0
   data8 0x3FC11111111111C2 // P1
   data8 0x3FABA1BA1BA0E850 // P2
   data8 0x3F9664F4886725A7 // P3
LOCAL_OBJECT_END(double_tan_constants)

LOCAL_OBJECT_START(double_Q_tan_constants)
   data8 0xC4C6628B80DC1CD1, 0x00003FBF // pi/2 lo
   data8 0x3E223A73BA576E48 // Q8
   data8 0x3DF54AD8D1F2CA43 // Q9
   data8 0x3EF66A8EE529A6AA // Q4
   data8 0x3EC2281050410EE6 // Q5
   data8 0x3E8D6BB992CC3CF5 // Q6
   data8 0x3E57F88DE34832E4 // Q7
   data8 0x3FD5555555555555 // Q0
   data8 0x3F96C16C16C16DB8 // Q1
   data8 0x3F61566ABBFFB489 // Q2
   data8 0x3F2BBD77945C1733 // Q3
   data8 0x3D927FB33E2B0E04 // Q10
LOCAL_OBJECT_END(double_Q_tan_constants)


.section .text

////////////////////////////////////////////////////////

LOCAL_LIBM_ENTRY(cot)
// The initial fnorm will take any unmasked faults and
// normalize any single/double unorms

{ .mlx
      cmp.eq    p12, p11 = r0, r0 // set p12=1, p11=0 for cot
      movl tan_GR_sig_inv_pi_by_2 = 0xA2F9836E4E44152A // significand of 2/pi
}
{ .mlx
      addl           tan_AD   = @ltoff(double_tan_constants), gp
      movl tan_GR_rshf_2to64 = 0x47e8000000000000 // 1.1000 2^(63+63+1)
}
;;

{ .mlx
      mov tan_GR_exp_2tom64 = 0xffff-64 // exponent of scaling factor 2^-64
      movl tan_GR_rshf = 0x43e8000000000000 // 1.1000 2^63 for right shift
}
{ .mfb
      ld8 tan_AD = [tan_AD]
      fnorm.s0  tan_NORM_f8  = f8
      br.cond.sptk COMMON_PATH
}
;;

LOCAL_LIBM_END(cot)

GLOBAL_IEEE754_ENTRY(tan)
// The initial fnorm will take any unmasked faults and
// normalize any single/double unorms

{ .mlx
      cmp.eq    p11, p12 = r0, r0 // set p11=1, p12=0 for tan
      movl tan_GR_sig_inv_pi_by_2 = 0xA2F9836E4E44152A // significand of 2/pi
}
{ .mlx
      addl           tan_AD   = @ltoff(double_tan_constants), gp
      movl tan_GR_rshf_2to64 = 0x47e8000000000000 // 1.1000 2^(63+63+1)
}
;;

{ .mlx
      mov tan_GR_exp_2tom64 = 0xffff-64 // exponent of scaling factor 2^-64
      movl tan_GR_rshf = 0x43e8000000000000 // 1.1000 2^63 for right shift
}
{ .mfi
      ld8 tan_AD = [tan_AD]
      fnorm.s0  tan_NORM_f8  = f8
      nop.i     0
}
;;


// Common path for both tan and cot
COMMON_PATH:
// Form two constants we need
//   2/pi * 2^1 * 2^63, scaled by 2^64 since we just loaded the significand
//   1.1000...000 * 2^(63+63+1) to right shift int(W) into the significand
{ .mmi
      setf.sig TAN_INV_PI_BY_2_2TO64 = tan_GR_sig_inv_pi_by_2
      setf.d TAN_RSHF_2TO64 = tan_GR_rshf_2to64
      mov       tan_GR_17_ones     = 0x1ffff             ;;
}


// Form another constant
//   2^-64 for scaling Nfloat
//   1.1000...000 * 2^63, the right shift constant
{ .mmf
      setf.exp TAN_2TOM64 = tan_GR_exp_2tom64
      adds tan_ADQ = double_Q_tan_constants - double_tan_constants, tan_AD
(p11) fclass.m.unc  p6,p0 = f8, 0x07  // Test for x=0 (tan)
}
;;


// Form another constant
//   2^-64 for scaling Nfloat
//   1.1000...000 * 2^63, the right shift constant
{ .mmf
      setf.d TAN_RSHF = tan_GR_rshf
      ldfe      tan_Pi_by_2_hi = [tan_AD],16
      fclass.m.unc  p7,p0 = f8, 0x23  // Test for x=inf
}
;;

{ .mfb
      ldfe      tan_Pi_by_2_lo = [tan_ADQ],16
      fclass.m.unc  p8,p0 = f8, 0xc3  // Test for x=nan
(p6)  br.ret.spnt    b0    ;;         // Exit for x=0 (tan only)
}

{ .mfi
      ldfpd     tan_P14,tan_P15 = [tan_AD],16
(p7)  frcpa.s0  f8,p9=f0,f0           // Set qnan indef if x=inf
      mov       tan_GR_10009 = 0x10009
}
{ .mib
      ldfpd      tan_Q8,tan_Q9  = [tan_ADQ],16
      nop.i 999
(p7)  br.ret.spnt    b0    ;;         // Exit for x=inf
}

{ .mfi
      ldfpd      tan_P12,tan_P13 = [tan_AD],16
(p12) fclass.m.unc  p6,p0 = f8, 0x07  // Test for x=0 (cot)
      nop.i 999
}
{ .mfb
      ldfpd      tan_Q4,tan_Q5  = [tan_ADQ],16
(p8)  fma.d.s0   f8=f8,f1,f8          // Set qnan if x=nan
(p8)  br.ret.spnt    b0    ;;         // Exit for x=nan
}

{ .mmf
      getf.exp  tan_signexp    = tan_NORM_f8
      ldfpd     tan_P8,tan_P9  = [tan_AD],16
      fmerge.s  arg_copy       = f8, f8 ;; // Save input for error call
}

// Multiply x by scaled 2/pi and add large const to shift integer part of W to
//   rightmost bits of significand
{ .mmf
      alloc      r32=ar.pfs,0,4,4,0
      ldfpd      tan_Q6,tan_Q7  = [tan_ADQ],16
      fma.s1 TAN_W_2TO64_RSH = tan_NORM_f8,TAN_INV_PI_BY_2_2TO64,TAN_RSHF_2TO64
};;

{ .mmf
      ldfpd      tan_P10,tan_P11 = [tan_AD],16
      and       tan_exp = tan_GR_17_ones, tan_signexp
(p6)  frcpa.s0  f8, p0 = f1, f8 ;;        // cot(+-0) = +-Inf
}


// p7 is true if we must call DBX TAN
// p7 is true if f8 exp is > 0x10009 (which includes all ones
//    NAN or inf)
{ .mmb
      ldfpd      tan_Q0,tan_Q1  = [tan_ADQ],16
      cmp.ge.unc  p7,p0 = tan_exp,tan_GR_10009
(p7)  br.cond.spnt   TAN_DBX ;;
}


{ .mmb
      ldfpd      tan_P4,tan_P5  = [tan_AD],16
(p6)  mov GR_Parameter_Tag = 226          // (cot)
(p6)  br.cond.spnt __libm_error_region ;; // call error support if cot(+-0)
}


{ .mmi
      ldfpd      tan_Q2,tan_Q3  = [tan_ADQ],16
      nop.m 999
      nop.i 999 ;;
}



// TAN_NFLOAT = Round_Int_Nearest(tan_W)
{ .mfi
      ldfpd      tan_P6,tan_P7  = [tan_AD],16
      fms.s1 TAN_NFLOAT = TAN_W_2TO64_RSH,TAN_2TOM64,TAN_RSHF
      nop.i 999 ;;
}


{ .mfi
      ldfd      tan_Q10 = [tan_ADQ]
      nop.f 999
      nop.i 999 ;;
}


{ .mfi
      ldfpd      tan_P0,tan_P1  = [tan_AD],16
      nop.f 999
      nop.i 999 ;;
}


{ .mmi
      getf.sig    tan_GR_n = TAN_W_2TO64_RSH
      ldfpd      tan_P2,tan_P3  = [tan_AD]
      nop.i 999 ;;
}

// tan_r          = -tan_Nfloat * tan_Pi_by_2_hi + x
{ .mfi
(p12) add        tan_GR_n = 0x1, tan_GR_n // N = N + 1 (for cot)
      fnma.s1  tan_r      = TAN_NFLOAT, tan_Pi_by_2_hi,  tan_NORM_f8
      nop.i 999 ;;
}


// p8 ==> even
// p9 ==> odd
{ .mmi
      and         tan_GR_N_odd_even = 0x1, tan_GR_n ;;
      nop.m 999
      cmp.eq.unc  p8,p9          = tan_GR_N_odd_even, r0      ;;
}


.pred.rel "mutex", p11, p12
// tan_r          = tan_r -tan_Nfloat * tan_Pi_by_2_lo (tan)
{ .mfi
      nop.m 999
(p11) fnma.s1  tan_r      = TAN_NFLOAT, tan_Pi_by_2_lo,  tan_r
      nop.i 999
}
// tan_r          = -(tan_r -tan_Nfloat * tan_Pi_by_2_lo) (cot)
{ .mfi
      nop.m 999
(p12) fms.s1   tan_r      = TAN_NFLOAT, tan_Pi_by_2_lo,  tan_r
      nop.i 999 ;;
}


{ .mfi
      nop.m 999
      fma.s1   tan_rsq    = tan_r, tan_r,   f0
      nop.i 999 ;;
}


{ .mfi
      nop.m 999
(p9)  frcpa.s1   tan_y0, p0 = f1,tan_r
      nop.i 999  ;;
}


{ .mfi
      nop.m 999
(p8)  fma.s1  tan_v18 = tan_rsq, tan_P15, tan_P14
      nop.i 999
}
{ .mfi
      nop.m 999
(p8)  fma.s1  tan_v4  = tan_rsq, tan_P1, tan_P0
      nop.i 999  ;;
}



{ .mfi
      nop.m 999
(p8)  fma.s1  tan_v16 = tan_rsq, tan_P13, tan_P12
      nop.i 999
}
{ .mfi
      nop.m 999
(p8)  fma.s1  tan_v17 = tan_rsq, tan_rsq, f0
      nop.i 999 ;;
}



{ .mfi
      nop.m 999
(p8)  fma.s1  tan_v12 = tan_rsq, tan_P9, tan_P8
      nop.i 999
}
{ .mfi
      nop.m 999
(p8)  fma.s1  tan_v13 = tan_rsq, tan_P11, tan_P10
      nop.i 999 ;;
}



{ .mfi
      nop.m 999
(p8)  fma.s1  tan_v7  = tan_rsq, tan_P5, tan_P4
      nop.i 999
}
{ .mfi
      nop.m 999
(p8)  fma.s1  tan_v8  = tan_rsq, tan_P7, tan_P6
      nop.i 999 ;;
}



{ .mfi
      nop.m 999
(p9)  fnma.s1    tan_d   = tan_r, tan_y0, f1
      nop.i 999
}
{ .mfi
      nop.m 999
(p8)  fma.s1  tan_v5  = tan_rsq, tan_P3, tan_P2
      nop.i 999 ;;
}



{ .mfi
      nop.m 999
(p9)  fma.s1  tan_z11 = tan_rsq, tan_Q9, tan_Q8
      nop.i 999
}
{ .mfi
      nop.m 999
(p9)  fma.s1  tan_z12 = tan_rsq, tan_rsq, f0
      nop.i 999 ;;
}


{ .mfi
      nop.m 999
(p8)  fma.s1  tan_v15 = tan_v17, tan_v18, tan_v16
      nop.i 999
}
{ .mfi
      nop.m 999
(p9)  fma.s1  tan_z7 = tan_rsq, tan_Q5, tan_Q4
      nop.i 999 ;;
}


{ .mfi
      nop.m 999
(p8)  fma.s1  tan_v11 = tan_v17, tan_v13, tan_v12
      nop.i 999
}
{ .mfi
      nop.m 999
(p9)  fma.s1  tan_z8 = tan_rsq, tan_Q7, tan_Q6
      nop.i 999 ;;
}



{ .mfi
      nop.m 999
(p8)  fma.s1  tan_v14 = tan_v17, tan_v17, f0
      nop.i 999
}
{ .mfi
      nop.m 999
(p9)  fma.s1  tan_z3 = tan_rsq, tan_Q1, tan_Q0
      nop.i 999 ;;
}




{ .mfi
      nop.m 999
(p8)  fma.s1  tan_v3 = tan_v17, tan_v5, tan_v4
      nop.i 999
}
{ .mfi
      nop.m 999
(p8)  fma.s1  tan_v6 = tan_v17, tan_v8, tan_v7
      nop.i 999 ;;
}



{ .mfi
      nop.m 999
(p9)  fma.s1     tan_y1  = tan_y0, tan_d, tan_y0
      nop.i 999
}
{ .mfi
      nop.m 999
(p9)  fma.s1     tan_dsq = tan_d, tan_d, f0
      nop.i 999 ;;
}


{ .mfi
      nop.m 999
(p9)  fma.s1  tan_z10 = tan_z12, tan_Q10, tan_z11
      nop.i 999
}
{ .mfi
      nop.m 999
(p9)  fma.s1  tan_z9  = tan_z12, tan_z12,f0
      nop.i 999 ;;
}


{ .mfi
      nop.m 999
(p9)  fma.s1  tan_z4 = tan_rsq, tan_Q3, tan_Q2
      nop.i 999
}
{ .mfi
      nop.m 999
(p9)  fma.s1  tan_z6  = tan_z12, tan_z8, tan_z7
      nop.i 999 ;;
}



{ .mfi
      nop.m 999
(p8)  fma.s1  tan_v10 = tan_v14, tan_v15, tan_v11
      nop.i 999 ;;
}



{ .mfi
      nop.m 999
(p9)  fma.s1     tan_y2  = tan_y1, tan_d, tan_y0
      nop.i 999
}
{ .mfi
      nop.m 999
(p9)  fma.s1     tan_d4  = tan_dsq, tan_dsq, tan_d
      nop.i 999  ;;
}


{ .mfi
      nop.m 999
(p8)  fma.s1  tan_v2 = tan_v14, tan_v6, tan_v3
      nop.i 999
}
{ .mfi
      nop.m 999
(p8)  fma.s1  tan_v9 = tan_v14, tan_v14, f0
      nop.i 999 ;;
}


{ .mfi
      nop.m 999
(p9)  fma.s1  tan_z2  = tan_z12, tan_z4, tan_z3
      nop.i 999
}
{ .mfi
      nop.m 999
(p9)  fma.s1  tan_z5  = tan_z9, tan_z10, tan_z6
      nop.i 999  ;;
}


{ .mfi
      nop.m 999
(p9)  fma.s1     tan_inv_r = tan_d4, tan_y2, tan_y0
      nop.i 999
}
{ .mfi
      nop.m 999
(p8)  fma.s1   tan_rcube  = tan_rsq, tan_r,   f0
      nop.i 999  ;;
}



{ .mfi
      nop.m 999
(p8)  fma.s1  tan_v1 = tan_v9, tan_v10, tan_v2
      nop.i 999
}
{ .mfi
      nop.m 999
(p9)  fma.s1  tan_z1  = tan_z9, tan_z5, tan_z2
      nop.i 999   ;;
}



{ .mfi
      nop.m 999
(p8)  fma.d.s0  f8  = tan_v1, tan_rcube, tan_r
      nop.i 999
}
{ .mfb
      nop.m 999
(p9)  fms.d.s0  f8  = tan_r, tan_z1, tan_inv_r
      br.ret.sptk    b0 ;;
}
GLOBAL_IEEE754_END(tan)

LOCAL_LIBM_ENTRY(__libm_callout)
TAN_DBX:
.prologue

{ .mfi
      nop.m 0
      fmerge.s f9 = f0,f0
.save ar.pfs,GR_SAVE_PFS
      mov  GR_SAVE_PFS=ar.pfs
}
;;

{ .mfi
      mov GR_SAVE_GP=gp
      nop.f 0
.save b0, GR_SAVE_B0
      mov GR_SAVE_B0=b0
}

.body
{ .mmb
      nop.m 999
      nop.m 999
(p11) br.cond.sptk.many  call_tanl ;;
}

// Here if we should call cotl
{ .mmb
      nop.m 999
      nop.m 999
      br.call.sptk.many  b0=__libm_cotl# ;;
}

{ .mfi
      mov gp        = GR_SAVE_GP
      fnorm.d.s0 f8 = f8
      mov b0        = GR_SAVE_B0
}
;;

{ .mib
      nop.m 999
      mov ar.pfs    = GR_SAVE_PFS
      br.ret.sptk     b0
;;
}

// Here if we should call tanl
call_tanl:
{ .mmb
      nop.m 999
      nop.m 999
      br.call.sptk.many  b0=__libm_tanl# ;;
}

{ .mfi
      mov gp        = GR_SAVE_GP
      fnorm.d.s0 f8 = f8
      mov b0        = GR_SAVE_B0
}
;;

{ .mib
      nop.m 999
      mov ar.pfs    = GR_SAVE_PFS
      br.ret.sptk     b0
;;
}

LOCAL_LIBM_END(__libm_callout)

.type __libm_tanl#,@function
.global __libm_tanl#
.type __libm_cotl#,@function
.global __libm_cotl#

LOCAL_LIBM_ENTRY(__libm_error_region)
.prologue

// (1)
{ .mfi
      add           GR_Parameter_Y=-32,sp        // Parameter 2 value
      nop.f 0
.save   ar.pfs,GR_SAVE_PFS
      mov           GR_SAVE_PFS=ar.pfs           // Save ar.pfs
}
{ .mfi
.fframe 64
      add sp=-64,sp                              // Create new stack
      nop.f 0
      mov GR_SAVE_GP=gp                          // Save gp
};;

// (2)
{ .mmi
      stfd [GR_Parameter_Y] = f1,16              // STORE Parameter 2 on stack
      add GR_Parameter_X = 16,sp                 // Parameter 1 address
.save   b0, GR_SAVE_B0
      mov GR_SAVE_B0=b0                          // Save b0
};;

.body
// (3)
{ .mib
      stfd [GR_Parameter_X] = arg_copy           // STORE Parameter 1 on stack
      add   GR_Parameter_RESULT = 0,GR_Parameter_Y  // Parameter 3 address
      nop.b 0
}
{ .mib
      stfd [GR_Parameter_Y] = f8                 // STORE Parameter 3 on stack
      add   GR_Parameter_Y = -16,GR_Parameter_Y
      br.call.sptk b0=__libm_error_support#      // Call error handling function
};;
{ .mmi
      nop.m 0
      nop.m 0
      add   GR_Parameter_RESULT = 48,sp
};;

// (4)
{ .mmi
      ldfd  f8 = [GR_Parameter_RESULT]           // Get return result off stack
.restore sp
      add   sp = 64,sp                           // Restore stack pointer
      mov   b0 = GR_SAVE_B0                      // Restore return address
};;
{ .mib
      mov   gp = GR_SAVE_GP                      // Restore gp
      mov   ar.pfs = GR_SAVE_PFS                 // Restore ar.pfs
      br.ret.sptk     b0                         // Return
};;

LOCAL_LIBM_END(__libm_error_region)

.type   __libm_error_support#,@function
.global __libm_error_support#