[J-u-boot.git] / arch / nios2 / lib / longlong.h

/* SPDX-License-Identifier: GPL-2.0+ */
/* longlong.h -- definitions for mixed size 32/64 bit arithmetic.
   Copyright (C) 1991, 1992, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2004,
   2005  Free Software Foundation, Inc.
 */

/* You have to define the following before including this file:

   UWtype -- An unsigned type, default type for operations (typically a "word")
   UHWtype -- An unsigned type, at least half the size of UWtype.
   UDWtype -- An unsigned type, at least twice as large a UWtype
   W_TYPE_SIZE -- size in bits of UWtype

   UQItype -- Unsigned 8 bit type.
   SItype, USItype -- Signed and unsigned 32 bit types.
   DItype, UDItype -- Signed and unsigned 64 bit types.

   On a 32 bit machine UWtype should typically be USItype;
   on a 64 bit machine, UWtype should typically be UDItype.  */

#define __BITS4 (W_TYPE_SIZE / 4)
#define __ll_B ((UWtype) 1 << (W_TYPE_SIZE / 2))
#define __ll_lowpart(t) ((UWtype) (t) & (__ll_B - 1))
#define __ll_highpart(t) ((UWtype) (t) >> (W_TYPE_SIZE / 2))

#ifndef W_TYPE_SIZE
#define W_TYPE_SIZE	32
#define UWtype		USItype
#define UHWtype		USItype
#define UDWtype		UDItype
#endif

extern const UQItype __clz_tab[256];

/* Define auxiliary asm macros.

   1) umul_ppmm(high_prod, low_prod, multiplier, multiplicand) multiplies two
   UWtype integers MULTIPLIER and MULTIPLICAND, and generates a two UWtype
   word product in HIGH_PROD and LOW_PROD.

   2) __umulsidi3(a,b) multiplies two UWtype integers A and B, and returns a
   UDWtype product.  This is just a variant of umul_ppmm.

   3) udiv_qrnnd(quotient, remainder, high_numerator, low_numerator,
   denominator) divides a UDWtype, composed by the UWtype integers
   HIGH_NUMERATOR and LOW_NUMERATOR, by DENOMINATOR and places the quotient
   in QUOTIENT and the remainder in REMAINDER.  HIGH_NUMERATOR must be less
   than DENOMINATOR for correct operation.  If, in addition, the most
   significant bit of DENOMINATOR must be 1, then the pre-processor symbol
   UDIV_NEEDS_NORMALIZATION is defined to 1.

   4) sdiv_qrnnd(quotient, remainder, high_numerator, low_numerator,
   denominator).  Like udiv_qrnnd but the numbers are signed.  The quotient
   is rounded towards 0.

   5) count_leading_zeros(count, x) counts the number of zero-bits from the
   msb to the first nonzero bit in the UWtype X.  This is the number of
   steps X needs to be shifted left to set the msb.  Undefined for X == 0,
   unless the symbol COUNT_LEADING_ZEROS_0 is defined to some value.

   6) count_trailing_zeros(count, x) like count_leading_zeros, but counts
   from the least significant end.

   7) add_ssaaaa(high_sum, low_sum, high_addend_1, low_addend_1,
   high_addend_2, low_addend_2) adds two UWtype integers, composed by
   HIGH_ADDEND_1 and LOW_ADDEND_1, and HIGH_ADDEND_2 and LOW_ADDEND_2
   respectively.  The result is placed in HIGH_SUM and LOW_SUM.  Overflow
   (i.e. carry out) is not stored anywhere, and is lost.

   8) sub_ddmmss(high_difference, low_difference, high_minuend, low_minuend,
   high_subtrahend, low_subtrahend) subtracts two two-word UWtype integers,
   composed by HIGH_MINUEND_1 and LOW_MINUEND_1, and HIGH_SUBTRAHEND_2 and
   LOW_SUBTRAHEND_2 respectively.  The result is placed in HIGH_DIFFERENCE
   and LOW_DIFFERENCE.  Overflow (i.e. carry out) is not stored anywhere,
   and is lost.

   If any of these macros are left undefined for a particular CPU,
   C macros are used.  */

/* The CPUs come in alphabetical order below.

   Please add support for more CPUs here, or improve the current support
   for the CPUs below!
   (E.g. WE32100, IBM360.)  */

/* Snipped per CPU support */

/* If this machine has no inline assembler, use C macros.  */

#if !defined (add_ssaaaa)
#define add_ssaaaa(sh, sl, ah, al, bh, bl) \
  do {									\
    UWtype __x;								\
    __x = (al) + (bl);							\
    (sh) = (ah) + (bh) + (__x < (al));					\
    (sl) = __x;								\
  } while (0)
#endif

#if !defined (sub_ddmmss)
#define sub_ddmmss(sh, sl, ah, al, bh, bl) \
  do {									\
    UWtype __x;								\
    __x = (al) - (bl);							\
    (sh) = (ah) - (bh) - (__x > (al));					\
    (sl) = __x;								\
  } while (0)
#endif

/* If we lack umul_ppmm but have smul_ppmm, define umul_ppmm in terms of
   smul_ppmm.  */
#if !defined (umul_ppmm) && defined (smul_ppmm)
#define umul_ppmm(w1, w0, u, v)						\
  do {									\
    UWtype __w1;							\
    UWtype __xm0 = (u), __xm1 = (v);					\
    smul_ppmm (__w1, w0, __xm0, __xm1);					\
    (w1) = __w1 + (-(__xm0 >> (W_TYPE_SIZE - 1)) & __xm1)		\
		+ (-(__xm1 >> (W_TYPE_SIZE - 1)) & __xm0);		\
  } while (0)
#endif

/* If we still don't have umul_ppmm, define it using plain C.  */
#if !defined (umul_ppmm)
#define umul_ppmm(w1, w0, u, v)						\
  do {									\
    UWtype __x0, __x1, __x2, __x3;					\
    UHWtype __ul, __vl, __uh, __vh;					\
									\
    __ul = __ll_lowpart (u);						\
    __uh = __ll_highpart (u);						\
    __vl = __ll_lowpart (v);						\
    __vh = __ll_highpart (v);						\
									\
    __x0 = (UWtype) __ul * __vl;					\
    __x1 = (UWtype) __ul * __vh;					\
    __x2 = (UWtype) __uh * __vl;					\
    __x3 = (UWtype) __uh * __vh;					\
									\
    __x1 += __ll_highpart (__x0);/* this can't give carry */		\
    __x1 += __x2;		/* but this indeed can */		\
    if (__x1 < __x2)		/* did we get it? */			\
      __x3 += __ll_B;		/* yes, add it in the proper pos.  */	\
									\
    (w1) = __x3 + __ll_highpart (__x1);					\
    (w0) = __ll_lowpart (__x1) * __ll_B + __ll_lowpart (__x0);		\
  } while (0)
#endif

#if !defined (__umulsidi3)
#define __umulsidi3(u, v) \
  ({DWunion __w;							\
    umul_ppmm (__w.s.high, __w.s.low, u, v);				\
    __w.ll; })
#endif

/* Define this unconditionally, so it can be used for debugging.  */
#define __udiv_qrnnd_c(q, r, n1, n0, d) \
  do {									\
    UWtype __d1, __d0, __q1, __q0;					\
    UWtype __r1, __r0, __m;						\
    __d1 = __ll_highpart (d);						\
    __d0 = __ll_lowpart (d);						\
									\
    __r1 = (n1) % __d1;							\
    __q1 = (n1) / __d1;							\
    __m = (UWtype) __q1 * __d0;						\
    __r1 = __r1 * __ll_B | __ll_highpart (n0);				\
    if (__r1 < __m)							\
      {									\
	__q1--, __r1 += (d);						\
	if (__r1 >= (d)) /* i.e. we didn't get carry when adding to __r1 */\
	  if (__r1 < __m)						\
	    __q1--, __r1 += (d);					\
      }									\
    __r1 -= __m;							\
									\
    __r0 = __r1 % __d1;							\
    __q0 = __r1 / __d1;							\
    __m = (UWtype) __q0 * __d0;						\
    __r0 = __r0 * __ll_B | __ll_lowpart (n0);				\
    if (__r0 < __m)							\
      {									\
	__q0--, __r0 += (d);						\
	if (__r0 >= (d))						\
	  if (__r0 < __m)						\
	    __q0--, __r0 += (d);					\
      }									\
    __r0 -= __m;							\
									\
    (q) = (UWtype) __q1 * __ll_B | __q0;				\
    (r) = __r0;								\
  } while (0)

/* If the processor has no udiv_qrnnd but sdiv_qrnnd, go through
   __udiv_w_sdiv (defined in libgcc or elsewhere).  */
#if !defined (udiv_qrnnd) && defined (sdiv_qrnnd)
#define udiv_qrnnd(q, r, nh, nl, d) \
  do {									\
    USItype __r;							\
    (q) = __udiv_w_sdiv (&__r, nh, nl, d);				\
    (r) = __r;								\
  } while (0)
#endif

/* If udiv_qrnnd was not defined for this processor, use __udiv_qrnnd_c.  */
#if !defined (udiv_qrnnd)
#define UDIV_NEEDS_NORMALIZATION 1
#define udiv_qrnnd __udiv_qrnnd_c
#endif

#if !defined (count_leading_zeros)
#define count_leading_zeros(count, x) \
  do {									\
    UWtype __xr = (x);							\
    UWtype __a;								\
									\
    if (W_TYPE_SIZE <= 32)						\
      {									\
	__a = __xr < ((UWtype)1<<2*__BITS4)				\
	  ? (__xr < ((UWtype)1<<__BITS4) ? 0 : __BITS4)			\
	  : (__xr < ((UWtype)1<<3*__BITS4) ?  2*__BITS4 : 3*__BITS4);	\
      }									\
    else								\
      {									\
	for (__a = W_TYPE_SIZE - 8; __a > 0; __a -= 8)			\
	  if (((__xr >> __a) & 0xff) != 0)				\
	    break;							\
      }									\
									\
    (count) = W_TYPE_SIZE - (__clz_tab[__xr >> __a] + __a);		\
  } while (0)
#define COUNT_LEADING_ZEROS_0 W_TYPE_SIZE
#endif

#if !defined (count_trailing_zeros)
/* Define count_trailing_zeros using count_leading_zeros.  The latter might be
   defined in asm, but if it is not, the C version above is good enough.  */
#define count_trailing_zeros(count, x) \
  do {									\
    UWtype __ctz_x = (x);						\
    UWtype __ctz_c;							\
    count_leading_zeros (__ctz_c, __ctz_x & -__ctz_x);			\
    (count) = W_TYPE_SIZE - 1 - __ctz_c;				\
  } while (0)
#endif

#ifndef UDIV_NEEDS_NORMALIZATION
#define UDIV_NEEDS_NORMALIZATION 0
#endif
Commit	Line	Data
83d290c5	1	/* SPDX-License-Identifier: GPL-2.0+ */
8d52ea6d TC	2	/* longlong.h -- definitions for mixed size 32/64 bit arithmetic.
	3	Copyright (C) 1991, 1992, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2004,
	4	2005 Free Software Foundation, Inc.
1a05b5f9	5	*/
8d52ea6d TC	6
	7	/* You have to define the following before including this file:
	8
	9	UWtype -- An unsigned type, default type for operations (typically a "word")
	10	UHWtype -- An unsigned type, at least half the size of UWtype.
	11	UDWtype -- An unsigned type, at least twice as large a UWtype
	12	W_TYPE_SIZE -- size in bits of UWtype
	13
	14	UQItype -- Unsigned 8 bit type.
	15	SItype, USItype -- Signed and unsigned 32 bit types.
	16	DItype, UDItype -- Signed and unsigned 64 bit types.
	17
	18	On a 32 bit machine UWtype should typically be USItype;
	19	on a 64 bit machine, UWtype should typically be UDItype. */
	20
	21	#define __BITS4 (W_TYPE_SIZE / 4)
	22	#define __ll_B ((UWtype) 1 << (W_TYPE_SIZE / 2))
	23	#define __ll_lowpart(t) ((UWtype) (t) & (__ll_B - 1))
	24	#define __ll_highpart(t) ((UWtype) (t) >> (W_TYPE_SIZE / 2))
	25
	26	#ifndef W_TYPE_SIZE
	27	#define W_TYPE_SIZE 32
	28	#define UWtype USItype
	29	#define UHWtype USItype
	30	#define UDWtype UDItype
	31	#endif
	32
	33	extern const UQItype __clz_tab[256];
	34
	35	/* Define auxiliary asm macros.
	36
	37	1) umul_ppmm(high_prod, low_prod, multiplier, multiplicand) multiplies two
	38	UWtype integers MULTIPLIER and MULTIPLICAND, and generates a two UWtype
	39	word product in HIGH_PROD and LOW_PROD.
	40
	41	2) __umulsidi3(a,b) multiplies two UWtype integers A and B, and returns a
	42	UDWtype product. This is just a variant of umul_ppmm.
	43
	44	3) udiv_qrnnd(quotient, remainder, high_numerator, low_numerator,
	45	denominator) divides a UDWtype, composed by the UWtype integers
	46	HIGH_NUMERATOR and LOW_NUMERATOR, by DENOMINATOR and places the quotient
	47	in QUOTIENT and the remainder in REMAINDER. HIGH_NUMERATOR must be less
	48	than DENOMINATOR for correct operation. If, in addition, the most
	49	significant bit of DENOMINATOR must be 1, then the pre-processor symbol
	50	UDIV_NEEDS_NORMALIZATION is defined to 1.
	51
	52	4) sdiv_qrnnd(quotient, remainder, high_numerator, low_numerator,
	53	denominator). Like udiv_qrnnd but the numbers are signed. The quotient
	54	is rounded towards 0.
	55
	56	5) count_leading_zeros(count, x) counts the number of zero-bits from the
	57	msb to the first nonzero bit in the UWtype X. This is the number of
	58	steps X needs to be shifted left to set the msb. Undefined for X == 0,
	59	unless the symbol COUNT_LEADING_ZEROS_0 is defined to some value.
	60
	61	6) count_trailing_zeros(count, x) like count_leading_zeros, but counts
	62	from the least significant end.
	63
	64	7) add_ssaaaa(high_sum, low_sum, high_addend_1, low_addend_1,
	65	high_addend_2, low_addend_2) adds two UWtype integers, composed by
	66	HIGH_ADDEND_1 and LOW_ADDEND_1, and HIGH_ADDEND_2 and LOW_ADDEND_2
	67	respectively. The result is placed in HIGH_SUM and LOW_SUM. Overflow
	68	(i.e. carry out) is not stored anywhere, and is lost.
	69
70	8) sub_ddmmss(high_difference, low_difference, high_minuend, low_minuend,
71	high_subtrahend, low_subtrahend) subtracts two two-word UWtype integers,
72	composed by HIGH_MINUEND_1 and LOW_MINUEND_1, and HIGH_SUBTRAHEND_2 and
73	LOW_SUBTRAHEND_2 respectively. The result is placed in HIGH_DIFFERENCE
74	and LOW_DIFFERENCE. Overflow (i.e. carry out) is not stored anywhere,
75	and is lost.
76
77	If any of these macros are left undefined for a particular CPU,
78	C macros are used. */
79
80	/* The CPUs come in alphabetical order below.
81
82	Please add support for more CPUs here, or improve the current support
83	for the CPUs below!
84	(E.g. WE32100, IBM360.) */
85
86	/* Snipped per CPU support */
87
88	/* If this machine has no inline assembler, use C macros. */
89
90	#if !defined (add_ssaaaa)
91	#define add_ssaaaa(sh, sl, ah, al, bh, bl) \
92	do { \
93	UWtype __x; \
94	__x = (al) + (bl); \
95	(sh) = (ah) + (bh) + (__x < (al)); \
96	(sl) = __x; \
97	} while (0)
98	#endif
99
100	#if !defined (sub_ddmmss)
101	#define sub_ddmmss(sh, sl, ah, al, bh, bl) \
102	do { \
103	UWtype __x; \
104	__x = (al) - (bl); \
105	(sh) = (ah) - (bh) - (__x > (al)); \
106	(sl) = __x; \
107	} while (0)
108	#endif
109
110	/* If we lack umul_ppmm but have smul_ppmm, define umul_ppmm in terms of
111	smul_ppmm. */
112	#if !defined (umul_ppmm) && defined (smul_ppmm)
113	#define umul_ppmm(w1, w0, u, v) \
114	do { \
115	UWtype __w1; \
116	UWtype __xm0 = (u), __xm1 = (v); \
117	smul_ppmm (__w1, w0, __xm0, __xm1); \
118	(w1) = __w1 + (-(__xm0 >> (W_TYPE_SIZE - 1)) & __xm1) \
119	+ (-(__xm1 >> (W_TYPE_SIZE - 1)) & __xm0); \
120	} while (0)
121	#endif
122
123	/* If we still don't have umul_ppmm, define it using plain C. */
124	#if !defined (umul_ppmm)
125	#define umul_ppmm(w1, w0, u, v) \
126	do { \
127	UWtype __x0, __x1, __x2, __x3; \
128	UHWtype __ul, __vl, __uh, __vh; \
129	\
130	__ul = __ll_lowpart (u); \
131	__uh = __ll_highpart (u); \
132	__vl = __ll_lowpart (v); \
133	__vh = __ll_highpart (v); \
134	\
135	__x0 = (UWtype) __ul * __vl; \
136	__x1 = (UWtype) __ul * __vh; \
137	__x2 = (UWtype) __uh * __vl; \
138	__x3 = (UWtype) __uh * __vh; \
139	\
140	__x1 += __ll_highpart (__x0);/* this can't give carry */ \
141	__x1 += __x2; /* but this indeed can */ \
142	if (__x1 < __x2) /* did we get it? */ \
143	__x3 += __ll_B; /* yes, add it in the proper pos. */ \
144	\
145	(w1) = __x3 + __ll_highpart (__x1); \
146	(w0) = __ll_lowpart (__x1) * __ll_B + __ll_lowpart (__x0); \
147	} while (0)
148	#endif
149
150	#if !defined (__umulsidi3)
151	#define __umulsidi3(u, v) \
152	({DWunion __w; \
153	umul_ppmm (__w.s.high, __w.s.low, u, v); \
154	__w.ll; })
155	#endif
156
157	/* Define this unconditionally, so it can be used for debugging. */
158	#define __udiv_qrnnd_c(q, r, n1, n0, d) \
159	do { \
160	UWtype __d1, __d0, __q1, __q0; \
161	UWtype __r1, __r0, __m; \
162	__d1 = __ll_highpart (d); \
163	__d0 = __ll_lowpart (d); \
164	\
165	__r1 = (n1) % __d1; \
166	__q1 = (n1) / __d1; \
167	__m = (UWtype) __q1 * __d0; \
168	__r1 = __r1 * __ll_B \| __ll_highpart (n0); \
169	if (__r1 < __m) \
170	{ \
171	__q1--, __r1 += (d); \
172	if (__r1 >= (d)) /* i.e. we didn't get carry when adding to __r1 */\
173	if (__r1 < __m) \
174	__q1--, __r1 += (d); \
175	} \
176	__r1 -= __m; \
177	\
178	__r0 = __r1 % __d1; \
179	__q0 = __r1 / __d1; \
180	__m = (UWtype) __q0 * __d0; \
181	__r0 = __r0 * __ll_B \| __ll_lowpart (n0); \
182	if (__r0 < __m) \
183	{ \
184	__q0--, __r0 += (d); \
185	if (__r0 >= (d)) \
186	if (__r0 < __m) \
187	__q0--, __r0 += (d); \
188	} \
189	__r0 -= __m; \
190	\
191	(q) = (UWtype) __q1 * __ll_B \| __q0; \
192	(r) = __r0; \
193	} while (0)
194
195	/* If the processor has no udiv_qrnnd but sdiv_qrnnd, go through
196	__udiv_w_sdiv (defined in libgcc or elsewhere). */
197	#if !defined (udiv_qrnnd) && defined (sdiv_qrnnd)
198	#define udiv_qrnnd(q, r, nh, nl, d) \
199	do { \
200	USItype __r; \
201	(q) = __udiv_w_sdiv (&__r, nh, nl, d); \
202	(r) = __r; \
203	} while (0)
204	#endif
205
206	/* If udiv_qrnnd was not defined for this processor, use __udiv_qrnnd_c. */
207	#if !defined (udiv_qrnnd)
208	#define UDIV_NEEDS_NORMALIZATION 1
209	#define udiv_qrnnd __udiv_qrnnd_c
210	#endif
211
212	#if !defined (count_leading_zeros)
213	#define count_leading_zeros(count, x) \
214	do { \
215	UWtype __xr = (x); \
216	UWtype __a; \
217	\
218	if (W_TYPE_SIZE <= 32) \
219	{ \
220	__a = __xr < ((UWtype)1<<2*__BITS4) \
221	? (__xr < ((UWtype)1<<__BITS4) ? 0 : __BITS4) \
222	: (__xr < ((UWtype)1<<3__BITS4) ? 2__BITS4 : 3*__BITS4); \
223	} \
224	else \
225	{ \
226	for (__a = W_TYPE_SIZE - 8; __a > 0; __a -= 8) \
227	if (((__xr >> __a) & 0xff) != 0) \
228	break; \
229	} \
230	\
231	(count) = W_TYPE_SIZE - (__clz_tab[__xr >> __a] + __a); \
232	} while (0)
233	#define COUNT_LEADING_ZEROS_0 W_TYPE_SIZE
234	#endif
235
236	#if !defined (count_trailing_zeros)
237	/* Define count_trailing_zeros using count_leading_zeros. The latter might be
238	defined in asm, but if it is not, the C version above is good enough. */
239	#define count_trailing_zeros(count, x) \
240	do { \
241	UWtype __ctz_x = (x); \
242	UWtype __ctz_c; \
243	count_leading_zeros (__ctz_c, __ctz_x & -__ctz_x); \
244	(count) = W_TYPE_SIZE - 1 - __ctz_c; \
245	} while (0)
246	#endif
247
248	#ifndef UDIV_NEEDS_NORMALIZATION
249	#define UDIV_NEEDS_NORMALIZATION 0
250	#endif