xref: /freebsd/lib/msun/src/math_private.h (revision 5ca8e32633c4ffbbcd6762e5888b6a4ba0708c6c)
1 /*
2  * ====================================================
3  * Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.
4  *
5  * Developed at SunPro, a Sun Microsystems, Inc. business.
6  * Permission to use, copy, modify, and distribute this
7  * software is freely granted, provided that this notice
8  * is preserved.
9  * ====================================================
10  */
11 
12 /*
13  */
14 
15 #ifndef _MATH_PRIVATE_H_
16 #define	_MATH_PRIVATE_H_
17 
18 #include <sys/types.h>
19 #include <machine/endian.h>
20 
21 /*
22  * The original fdlibm code used statements like:
23  *	n0 = ((*(int*)&one)>>29)^1;		* index of high word *
24  *	ix0 = *(n0+(int*)&x);			* high word of x *
25  *	ix1 = *((1-n0)+(int*)&x);		* low word of x *
26  * to dig two 32 bit words out of the 64 bit IEEE floating point
27  * value.  That is non-ANSI, and, moreover, the gcc instruction
28  * scheduler gets it wrong.  We instead use the following macros.
29  * Unlike the original code, we determine the endianness at compile
30  * time, not at run time; I don't see much benefit to selecting
31  * endianness at run time.
32  */
33 
34 /*
35  * A union which permits us to convert between a double and two 32 bit
36  * ints.
37  */
38 
39 #ifdef __arm__
40 #if defined(__VFP_FP__) || defined(__ARM_EABI__)
41 #define	IEEE_WORD_ORDER	BYTE_ORDER
42 #else
43 #define	IEEE_WORD_ORDER	BIG_ENDIAN
44 #endif
45 #else /* __arm__ */
46 #define	IEEE_WORD_ORDER	BYTE_ORDER
47 #endif
48 
49 /* A union which permits us to convert between a long double and
50    four 32 bit ints.  */
51 
52 #if IEEE_WORD_ORDER == BIG_ENDIAN
53 
54 typedef union
55 {
56   long double value;
57   struct {
58     u_int32_t mswhi;
59     u_int32_t mswlo;
60     u_int32_t lswhi;
61     u_int32_t lswlo;
62   } parts32;
63   struct {
64     u_int64_t msw;
65     u_int64_t lsw;
66   } parts64;
67 } ieee_quad_shape_type;
68 
69 #endif
70 
71 #if IEEE_WORD_ORDER == LITTLE_ENDIAN
72 
73 typedef union
74 {
75   long double value;
76   struct {
77     u_int32_t lswlo;
78     u_int32_t lswhi;
79     u_int32_t mswlo;
80     u_int32_t mswhi;
81   } parts32;
82   struct {
83     u_int64_t lsw;
84     u_int64_t msw;
85   } parts64;
86 } ieee_quad_shape_type;
87 
88 #endif
89 
90 #if IEEE_WORD_ORDER == BIG_ENDIAN
91 
92 typedef union
93 {
94   double value;
95   struct
96   {
97     u_int32_t msw;
98     u_int32_t lsw;
99   } parts;
100   struct
101   {
102     u_int64_t w;
103   } xparts;
104 } ieee_double_shape_type;
105 
106 #endif
107 
108 #if IEEE_WORD_ORDER == LITTLE_ENDIAN
109 
110 typedef union
111 {
112   double value;
113   struct
114   {
115     u_int32_t lsw;
116     u_int32_t msw;
117   } parts;
118   struct
119   {
120     u_int64_t w;
121   } xparts;
122 } ieee_double_shape_type;
123 
124 #endif
125 
126 /* Get two 32 bit ints from a double.  */
127 
128 #define EXTRACT_WORDS(ix0,ix1,d)				\
129 do {								\
130   ieee_double_shape_type ew_u;					\
131   ew_u.value = (d);						\
132   (ix0) = ew_u.parts.msw;					\
133   (ix1) = ew_u.parts.lsw;					\
134 } while (0)
135 
136 /* Get a 64-bit int from a double. */
137 #define EXTRACT_WORD64(ix,d)					\
138 do {								\
139   ieee_double_shape_type ew_u;					\
140   ew_u.value = (d);						\
141   (ix) = ew_u.xparts.w;						\
142 } while (0)
143 
144 /* Get the more significant 32 bit int from a double.  */
145 
146 #define GET_HIGH_WORD(i,d)					\
147 do {								\
148   ieee_double_shape_type gh_u;					\
149   gh_u.value = (d);						\
150   (i) = gh_u.parts.msw;						\
151 } while (0)
152 
153 /* Get the less significant 32 bit int from a double.  */
154 
155 #define GET_LOW_WORD(i,d)					\
156 do {								\
157   ieee_double_shape_type gl_u;					\
158   gl_u.value = (d);						\
159   (i) = gl_u.parts.lsw;						\
160 } while (0)
161 
162 /* Set a double from two 32 bit ints.  */
163 
164 #define INSERT_WORDS(d,ix0,ix1)					\
165 do {								\
166   ieee_double_shape_type iw_u;					\
167   iw_u.parts.msw = (ix0);					\
168   iw_u.parts.lsw = (ix1);					\
169   (d) = iw_u.value;						\
170 } while (0)
171 
172 /* Set a double from a 64-bit int. */
173 #define INSERT_WORD64(d,ix)					\
174 do {								\
175   ieee_double_shape_type iw_u;					\
176   iw_u.xparts.w = (ix);						\
177   (d) = iw_u.value;						\
178 } while (0)
179 
180 /* Set the more significant 32 bits of a double from an int.  */
181 
182 #define SET_HIGH_WORD(d,v)					\
183 do {								\
184   ieee_double_shape_type sh_u;					\
185   sh_u.value = (d);						\
186   sh_u.parts.msw = (v);						\
187   (d) = sh_u.value;						\
188 } while (0)
189 
190 /* Set the less significant 32 bits of a double from an int.  */
191 
192 #define SET_LOW_WORD(d,v)					\
193 do {								\
194   ieee_double_shape_type sl_u;					\
195   sl_u.value = (d);						\
196   sl_u.parts.lsw = (v);						\
197   (d) = sl_u.value;						\
198 } while (0)
199 
200 /*
201  * A union which permits us to convert between a float and a 32 bit
202  * int.
203  */
204 
205 typedef union
206 {
207   float value;
208   /* FIXME: Assumes 32 bit int.  */
209   unsigned int word;
210 } ieee_float_shape_type;
211 
212 /* Get a 32 bit int from a float.  */
213 
214 #define GET_FLOAT_WORD(i,d)					\
215 do {								\
216   ieee_float_shape_type gf_u;					\
217   gf_u.value = (d);						\
218   (i) = gf_u.word;						\
219 } while (0)
220 
221 /* Set a float from a 32 bit int.  */
222 
223 #define SET_FLOAT_WORD(d,i)					\
224 do {								\
225   ieee_float_shape_type sf_u;					\
226   sf_u.word = (i);						\
227   (d) = sf_u.value;						\
228 } while (0)
229 
230 /*
231  * Get expsign and mantissa as 16 bit and 64 bit ints from an 80 bit long
232  * double.
233  */
234 
235 #define	EXTRACT_LDBL80_WORDS(ix0,ix1,d)				\
236 do {								\
237   union IEEEl2bits ew_u;					\
238   ew_u.e = (d);							\
239   (ix0) = ew_u.xbits.expsign;					\
240   (ix1) = ew_u.xbits.man;					\
241 } while (0)
242 
243 /*
244  * Get expsign and mantissa as one 16 bit and two 64 bit ints from a 128 bit
245  * long double.
246  */
247 
248 #define	EXTRACT_LDBL128_WORDS(ix0,ix1,ix2,d)			\
249 do {								\
250   union IEEEl2bits ew_u;					\
251   ew_u.e = (d);							\
252   (ix0) = ew_u.xbits.expsign;					\
253   (ix1) = ew_u.xbits.manh;					\
254   (ix2) = ew_u.xbits.manl;					\
255 } while (0)
256 
257 /* Get expsign as a 16 bit int from a long double.  */
258 
259 #define	GET_LDBL_EXPSIGN(i,d)					\
260 do {								\
261   union IEEEl2bits ge_u;					\
262   ge_u.e = (d);							\
263   (i) = ge_u.xbits.expsign;					\
264 } while (0)
265 
266 /*
267  * Set an 80 bit long double from a 16 bit int expsign and a 64 bit int
268  * mantissa.
269  */
270 
271 #define	INSERT_LDBL80_WORDS(d,ix0,ix1)				\
272 do {								\
273   union IEEEl2bits iw_u;					\
274   iw_u.xbits.expsign = (ix0);					\
275   iw_u.xbits.man = (ix1);					\
276   (d) = iw_u.e;							\
277 } while (0)
278 
279 /*
280  * Set a 128 bit long double from a 16 bit int expsign and two 64 bit ints
281  * comprising the mantissa.
282  */
283 
284 #define	INSERT_LDBL128_WORDS(d,ix0,ix1,ix2)			\
285 do {								\
286   union IEEEl2bits iw_u;					\
287   iw_u.xbits.expsign = (ix0);					\
288   iw_u.xbits.manh = (ix1);					\
289   iw_u.xbits.manl = (ix2);					\
290   (d) = iw_u.e;							\
291 } while (0)
292 
293 /* Set expsign of a long double from a 16 bit int.  */
294 
295 #define	SET_LDBL_EXPSIGN(d,v)					\
296 do {								\
297   union IEEEl2bits se_u;					\
298   se_u.e = (d);							\
299   se_u.xbits.expsign = (v);					\
300   (d) = se_u.e;							\
301 } while (0)
302 
303 #ifdef __i386__
304 /* Long double constants are broken on i386. */
305 #define	LD80C(m, ex, v) {						\
306 	.xbits.man = __CONCAT(m, ULL),					\
307 	.xbits.expsign = (0x3fff + (ex)) | ((v) < 0 ? 0x8000 : 0),	\
308 }
309 #else
310 /* The above works on non-i386 too, but we use this to check v. */
311 #define	LD80C(m, ex, v)	{ .e = (v), }
312 #endif
313 
314 #ifdef FLT_EVAL_METHOD
315 /*
316  * Attempt to get strict C99 semantics for assignment with non-C99 compilers.
317  */
318 #if FLT_EVAL_METHOD == 0 || __GNUC__ == 0
319 #define	STRICT_ASSIGN(type, lval, rval)	((lval) = (rval))
320 #else
321 #define	STRICT_ASSIGN(type, lval, rval) do {	\
322 	volatile type __lval;			\
323 						\
324 	if (sizeof(type) >= sizeof(long double))	\
325 		(lval) = (rval);		\
326 	else {					\
327 		__lval = (rval);		\
328 		(lval) = __lval;		\
329 	}					\
330 } while (0)
331 #endif
332 #endif /* FLT_EVAL_METHOD */
333 
334 /* Support switching the mode to FP_PE if necessary. */
335 #if defined(__i386__) && !defined(NO_FPSETPREC)
336 #define	ENTERI() ENTERIT(long double)
337 #define	ENTERIT(returntype)			\
338 	returntype __retval;			\
339 	fp_prec_t __oprec;			\
340 						\
341 	if ((__oprec = fpgetprec()) != FP_PE)	\
342 		fpsetprec(FP_PE)
343 #define	RETURNI(x) do {				\
344 	__retval = (x);				\
345 	if (__oprec != FP_PE)			\
346 		fpsetprec(__oprec);		\
347 	RETURNF(__retval);			\
348 } while (0)
349 #define	ENTERV()				\
350 	fp_prec_t __oprec;			\
351 						\
352 	if ((__oprec = fpgetprec()) != FP_PE)	\
353 		fpsetprec(FP_PE)
354 #define	RETURNV() do {				\
355 	if (__oprec != FP_PE)			\
356 		fpsetprec(__oprec);		\
357 	return;			\
358 } while (0)
359 #else
360 #define	ENTERI()
361 #define	ENTERIT(x)
362 #define	RETURNI(x)	RETURNF(x)
363 #define	ENTERV()
364 #define	RETURNV()	return
365 #endif
366 
367 /* Default return statement if hack*_t() is not used. */
368 #define      RETURNF(v)      return (v)
369 
370 /*
371  * 2sum gives the same result as 2sumF without requiring |a| >= |b| or
372  * a == 0, but is slower.
373  */
374 #define	_2sum(a, b) do {	\
375 	__typeof(a) __s, __w;	\
376 				\
377 	__w = (a) + (b);	\
378 	__s = __w - (a);	\
379 	(b) = ((a) - (__w - __s)) + ((b) - __s); \
380 	(a) = __w;		\
381 } while (0)
382 
383 /*
384  * 2sumF algorithm.
385  *
386  * "Normalize" the terms in the infinite-precision expression a + b for
387  * the sum of 2 floating point values so that b is as small as possible
388  * relative to 'a'.  (The resulting 'a' is the value of the expression in
389  * the same precision as 'a' and the resulting b is the rounding error.)
390  * |a| must be >= |b| or 0, b's type must be no larger than 'a's type, and
391  * exponent overflow or underflow must not occur.  This uses a Theorem of
392  * Dekker (1971).  See Knuth (1981) 4.2.2 Theorem C.  The name "TwoSum"
393  * is apparently due to Skewchuk (1997).
394  *
395  * For this to always work, assignment of a + b to 'a' must not retain any
396  * extra precision in a + b.  This is required by C standards but broken
397  * in many compilers.  The brokenness cannot be worked around using
398  * STRICT_ASSIGN() like we do elsewhere, since the efficiency of this
399  * algorithm would be destroyed by non-null strict assignments.  (The
400  * compilers are correct to be broken -- the efficiency of all floating
401  * point code calculations would be destroyed similarly if they forced the
402  * conversions.)
403  *
404  * Fortunately, a case that works well can usually be arranged by building
405  * any extra precision into the type of 'a' -- 'a' should have type float_t,
406  * double_t or long double.  b's type should be no larger than 'a's type.
407  * Callers should use these types with scopes as large as possible, to
408  * reduce their own extra-precision and efficiciency problems.  In
409  * particular, they shouldn't convert back and forth just to call here.
410  */
411 #ifdef DEBUG
412 #define	_2sumF(a, b) do {				\
413 	__typeof(a) __w;				\
414 	volatile __typeof(a) __ia, __ib, __r, __vw;	\
415 							\
416 	__ia = (a);					\
417 	__ib = (b);					\
418 	assert(__ia == 0 || fabsl(__ia) >= fabsl(__ib));	\
419 							\
420 	__w = (a) + (b);				\
421 	(b) = ((a) - __w) + (b);			\
422 	(a) = __w;					\
423 							\
424 	/* The next 2 assertions are weak if (a) is already long double. */ \
425 	assert((long double)__ia + __ib == (long double)(a) + (b));	\
426 	__vw = __ia + __ib;				\
427 	__r = __ia - __vw;				\
428 	__r += __ib;					\
429 	assert(__vw == (a) && __r == (b));		\
430 } while (0)
431 #else /* !DEBUG */
432 #define	_2sumF(a, b) do {	\
433 	__typeof(a) __w;	\
434 				\
435 	__w = (a) + (b);	\
436 	(b) = ((a) - __w) + (b); \
437 	(a) = __w;		\
438 } while (0)
439 #endif /* DEBUG */
440 
441 /*
442  * Set x += c, where x is represented in extra precision as a + b.
443  * x must be sufficiently normalized and sufficiently larger than c,
444  * and the result is then sufficiently normalized.
445  *
446  * The details of ordering are that |a| must be >= |c| (so that (a, c)
447  * can be normalized without extra work to swap 'a' with c).  The details of
448  * the normalization are that b must be small relative to the normalized 'a'.
449  * Normalization of (a, c) makes the normalized c tiny relative to the
450  * normalized a, so b remains small relative to 'a' in the result.  However,
451  * b need not ever be tiny relative to 'a'.  For example, b might be about
452  * 2**20 times smaller than 'a' to give about 20 extra bits of precision.
453  * That is usually enough, and adding c (which by normalization is about
454  * 2**53 times smaller than a) cannot change b significantly.  However,
455  * cancellation of 'a' with c in normalization of (a, c) may reduce 'a'
456  * significantly relative to b.  The caller must ensure that significant
457  * cancellation doesn't occur, either by having c of the same sign as 'a',
458  * or by having |c| a few percent smaller than |a|.  Pre-normalization of
459  * (a, b) may help.
460  *
461  * This is a variant of an algorithm of Kahan (see Knuth (1981) 4.2.2
462  * exercise 19).  We gain considerable efficiency by requiring the terms to
463  * be sufficiently normalized and sufficiently increasing.
464  */
465 #define	_3sumF(a, b, c) do {	\
466 	__typeof(a) __tmp;	\
467 				\
468 	__tmp = (c);		\
469 	_2sumF(__tmp, (a));	\
470 	(b) += (a);		\
471 	(a) = __tmp;		\
472 } while (0)
473 
474 /*
475  * Common routine to process the arguments to nan(), nanf(), and nanl().
476  */
477 void _scan_nan(uint32_t *__words, int __num_words, const char *__s);
478 
479 /*
480  * Mix 0, 1 or 2 NaNs.  First add 0 to each arg.  This normally just turns
481  * signaling NaNs into quiet NaNs by setting a quiet bit.  We do this
482  * because we want to never return a signaling NaN, and also because we
483  * don't want the quiet bit to affect the result.  Then mix the converted
484  * args using the specified operation.
485  *
486  * When one arg is NaN, the result is typically that arg quieted.  When both
487  * args are NaNs, the result is typically the quietening of the arg whose
488  * mantissa is largest after quietening.  When neither arg is NaN, the
489  * result may be NaN because it is indeterminate, or finite for subsequent
490  * construction of a NaN as the indeterminate 0.0L/0.0L.
491  *
492  * Technical complications: the result in bits after rounding to the final
493  * precision might depend on the runtime precision and/or on compiler
494  * optimizations, especially when different register sets are used for
495  * different precisions.  Try to make the result not depend on at least the
496  * runtime precision by always doing the main mixing step in long double
497  * precision.  Try to reduce dependencies on optimizations by adding the
498  * the 0's in different precisions (unless everything is in long double
499  * precision).
500  */
501 #define	nan_mix(x, y)		(nan_mix_op((x), (y), +))
502 #define	nan_mix_op(x, y, op)	(((x) + 0.0L) op ((y) + 0))
503 
504 #ifdef _COMPLEX_H
505 
506 /*
507  * C99 specifies that complex numbers have the same representation as
508  * an array of two elements, where the first element is the real part
509  * and the second element is the imaginary part.
510  */
511 typedef union {
512 	float complex f;
513 	float a[2];
514 } float_complex;
515 typedef union {
516 	double complex f;
517 	double a[2];
518 } double_complex;
519 typedef union {
520 	long double complex f;
521 	long double a[2];
522 } long_double_complex;
523 #define	REALPART(z)	((z).a[0])
524 #define	IMAGPART(z)	((z).a[1])
525 
526 /*
527  * Inline functions that can be used to construct complex values.
528  *
529  * The C99 standard intends x+I*y to be used for this, but x+I*y is
530  * currently unusable in general since gcc introduces many overflow,
531  * underflow, sign and efficiency bugs by rewriting I*y as
532  * (0.0+I)*(y+0.0*I) and laboriously computing the full complex product.
533  * In particular, I*Inf is corrupted to NaN+I*Inf, and I*-0 is corrupted
534  * to -0.0+I*0.0.
535  *
536  * The C11 standard introduced the macros CMPLX(), CMPLXF() and CMPLXL()
537  * to construct complex values.  Compilers that conform to the C99
538  * standard require the following functions to avoid the above issues.
539  */
540 
541 #ifndef CMPLXF
542 static __inline float complex
543 CMPLXF(float x, float y)
544 {
545 	float_complex z;
546 
547 	REALPART(z) = x;
548 	IMAGPART(z) = y;
549 	return (z.f);
550 }
551 #endif
552 
553 #ifndef CMPLX
554 static __inline double complex
555 CMPLX(double x, double y)
556 {
557 	double_complex z;
558 
559 	REALPART(z) = x;
560 	IMAGPART(z) = y;
561 	return (z.f);
562 }
563 #endif
564 
565 #ifndef CMPLXL
566 static __inline long double complex
567 CMPLXL(long double x, long double y)
568 {
569 	long_double_complex z;
570 
571 	REALPART(z) = x;
572 	IMAGPART(z) = y;
573 	return (z.f);
574 }
575 #endif
576 
577 #endif /* _COMPLEX_H */
578 
579 /*
580  * The rnint() family rounds to the nearest integer for a restricted range
581  * range of args (up to about 2**MANT_DIG).  We assume that the current
582  * rounding mode is FE_TONEAREST so that this can be done efficiently.
583  * Extra precision causes more problems in practice, and we only centralize
584  * this here to reduce those problems, and have not solved the efficiency
585  * problems.  The exp2() family uses a more delicate version of this that
586  * requires extracting bits from the intermediate value, so it is not
587  * centralized here and should copy any solution of the efficiency problems.
588  */
589 
590 static inline double
591 rnint(__double_t x)
592 {
593 	/*
594 	 * This casts to double to kill any extra precision.  This depends
595 	 * on the cast being applied to a double_t to avoid compiler bugs
596 	 * (this is a cleaner version of STRICT_ASSIGN()).  This is
597 	 * inefficient if there actually is extra precision, but is hard
598 	 * to improve on.  We use double_t in the API to minimise conversions
599 	 * for just calling here.  Note that we cannot easily change the
600 	 * magic number to the one that works directly with double_t, since
601 	 * the rounding precision is variable at runtime on x86 so the
602 	 * magic number would need to be variable.  Assuming that the
603 	 * rounding precision is always the default is too fragile.  This
604 	 * and many other complications will move when the default is
605 	 * changed to FP_PE.
606 	 */
607 	return ((double)(x + 0x1.8p52) - 0x1.8p52);
608 }
609 
610 static inline float
611 rnintf(__float_t x)
612 {
613 	/*
614 	 * As for rnint(), except we could just call that to handle the
615 	 * extra precision case, usually without losing efficiency.
616 	 */
617 	return ((float)(x + 0x1.8p23F) - 0x1.8p23F);
618 }
619 
620 #ifdef LDBL_MANT_DIG
621 /*
622  * The complications for extra precision are smaller for rnintl() since it
623  * can safely assume that the rounding precision has been increased from
624  * its default to FP_PE on x86.  We don't exploit that here to get small
625  * optimizations from limiting the range to double.  We just need it for
626  * the magic number to work with long doubles.  ld128 callers should use
627  * rnint() instead of this if possible.  ld80 callers should prefer
628  * rnintl() since for amd64 this avoids swapping the register set, while
629  * for i386 it makes no difference (assuming FP_PE), and for other arches
630  * it makes little difference.
631  */
632 static inline long double
633 rnintl(long double x)
634 {
635 	return (x + __CONCAT(0x1.8p, LDBL_MANT_DIG) / 2 -
636 	    __CONCAT(0x1.8p, LDBL_MANT_DIG) / 2);
637 }
638 #endif /* LDBL_MANT_DIG */
639 
640 /*
641  * irint() and i64rint() give the same result as casting to their integer
642  * return type provided their arg is a floating point integer.  They can
643  * sometimes be more efficient because no rounding is required.
644  */
645 #if defined(amd64) || defined(__i386__)
646 #define	irint(x)						\
647     (sizeof(x) == sizeof(float) &&				\
648     sizeof(__float_t) == sizeof(long double) ? irintf(x) :	\
649     sizeof(x) == sizeof(double) &&				\
650     sizeof(__double_t) == sizeof(long double) ? irintd(x) :	\
651     sizeof(x) == sizeof(long double) ? irintl(x) : (int)(x))
652 #else
653 #define	irint(x)	((int)(x))
654 #endif
655 
656 #define	i64rint(x)	((int64_t)(x))	/* only needed for ld128 so not opt. */
657 
658 #if defined(__i386__)
659 static __inline int
660 irintf(float x)
661 {
662 	int n;
663 
664 	__asm("fistl %0" : "=m" (n) : "t" (x));
665 	return (n);
666 }
667 
668 static __inline int
669 irintd(double x)
670 {
671 	int n;
672 
673 	__asm("fistl %0" : "=m" (n) : "t" (x));
674 	return (n);
675 }
676 #endif
677 
678 #if defined(__amd64__) || defined(__i386__)
679 static __inline int
680 irintl(long double x)
681 {
682 	int n;
683 
684 	__asm("fistl %0" : "=m" (n) : "t" (x));
685 	return (n);
686 }
687 #endif
688 
689 /*
690  * The following are fast floor macros for 0 <= |x| < 0x1p(N-1), where
691  * N is the precision of the type of x. These macros are used in the
692  * half-cycle trignometric functions (e.g., sinpi(x)).
693  */
694 #define	FFLOORF(x, j0, ix) do {			\
695 	(j0) = (((ix) >> 23) & 0xff) - 0x7f;	\
696 	(ix) &= ~(0x007fffff >> (j0));		\
697 	SET_FLOAT_WORD((x), (ix));		\
698 } while (0)
699 
700 #define	FFLOOR(x, j0, ix, lx) do {				\
701 	(j0) = (((ix) >> 20) & 0x7ff) - 0x3ff;			\
702 	if ((j0) < 20) {					\
703 		(ix) &= ~(0x000fffff >> (j0));			\
704 		(lx) = 0;					\
705 	} else {						\
706 		(lx) &= ~((uint32_t)0xffffffff >> ((j0) - 20));	\
707 	}							\
708 	INSERT_WORDS((x), (ix), (lx));				\
709 } while (0)
710 
711 #define	FFLOORL80(x, j0, ix, lx) do {			\
712 	j0 = ix - 0x3fff + 1;				\
713 	if ((j0) < 32) {				\
714 		(lx) = ((lx) >> 32) << 32;		\
715 		(lx) &= ~((((lx) << 32)-1) >> (j0));	\
716 	} else {					\
717 		uint64_t _m;				\
718 		_m = (uint64_t)-1 >> (j0);		\
719 		if ((lx) & _m) (lx) &= ~_m;		\
720 	}						\
721 	INSERT_LDBL80_WORDS((x), (ix), (lx));		\
722 } while (0)
723 
724 #define FFLOORL128(x, ai, ar) do {			\
725 	union IEEEl2bits u;				\
726 	uint64_t m;					\
727 	int e;						\
728 	u.e = (x);					\
729 	e = u.bits.exp - 16383;				\
730 	if (e < 48) {					\
731 		m = ((1llu << 49) - 1) >> (e + 1);	\
732 		u.bits.manh &= ~m;			\
733 		u.bits.manl = 0;			\
734 	} else {					\
735 		m = (uint64_t)-1 >> (e - 48);		\
736 		u.bits.manl &= ~m;			\
737 	}						\
738 	(ai) = u.e;					\
739 	(ar) = (x) - (ai);				\
740 } while (0)
741 
742 #ifdef DEBUG
743 #if defined(__amd64__) || defined(__i386__)
744 #define	breakpoint()	asm("int $3")
745 #else
746 #include <signal.h>
747 
748 #define	breakpoint()	raise(SIGTRAP)
749 #endif
750 #endif
751 
752 #ifdef STRUCT_RETURN
753 #define	RETURNSP(rp) do {		\
754 	if (!(rp)->lo_set)		\
755 		RETURNF((rp)->hi);	\
756 	RETURNF((rp)->hi + (rp)->lo);	\
757 } while (0)
758 #define	RETURNSPI(rp) do {		\
759 	if (!(rp)->lo_set)		\
760 		RETURNI((rp)->hi);	\
761 	RETURNI((rp)->hi + (rp)->lo);	\
762 } while (0)
763 #endif
764 
765 #define	SUM2P(x, y) ({			\
766 	const __typeof (x) __x = (x);	\
767 	const __typeof (y) __y = (y);	\
768 	__x + __y;			\
769 })
770 
771 /* fdlibm kernel function */
772 int	__kernel_rem_pio2(double*,double*,int,int,int);
773 
774 /* double precision kernel functions */
775 #ifndef INLINE_REM_PIO2
776 int	__ieee754_rem_pio2(double,double*);
777 #endif
778 double	__kernel_sin(double,double,int);
779 double	__kernel_cos(double,double);
780 double	__kernel_tan(double,double,int);
781 double	__ldexp_exp(double,int);
782 #ifdef _COMPLEX_H
783 double complex __ldexp_cexp(double complex,int);
784 #endif
785 
786 /* float precision kernel functions */
787 #ifndef INLINE_REM_PIO2F
788 int	__ieee754_rem_pio2f(float,double*);
789 #endif
790 #ifndef INLINE_KERNEL_SINDF
791 float	__kernel_sindf(double);
792 #endif
793 #ifndef INLINE_KERNEL_COSDF
794 float	__kernel_cosdf(double);
795 #endif
796 #ifndef INLINE_KERNEL_TANDF
797 float	__kernel_tandf(double,int);
798 #endif
799 float	__ldexp_expf(float,int);
800 #ifdef _COMPLEX_H
801 float complex __ldexp_cexpf(float complex,int);
802 #endif
803 
804 /* long double precision kernel functions */
805 long double __kernel_sinl(long double, long double, int);
806 long double __kernel_cosl(long double, long double);
807 long double __kernel_tanl(long double, long double, int);
808 
809 #endif /* !_MATH_PRIVATE_H_ */
810