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
|
/* -*- linux-c -*- */
/* Math functions
* Copyright (C) 2005 Red Hat Inc.
* Portions (C) Free Software Foundation, Inc.
*
* This file is part of systemtap, and is free software. You can
* redistribute it and/or modify it under the terms of the GNU General
* Public License (GPL); either version 2, or (at your option) any
* later version.
*/
#ifndef _ARITH_C_
#define _ARITH_C_
/** @file arith.
* @brief Implements various arithmetic-related helper functions
*/
/* 64-bit division for 64-bit cpus and i386 */
/* Other 32-bit cpus will need to modify this file. */
#ifdef __i386__
long long _div64 (long long u, long long v);
long long _mod64 (long long u, long long v);
#endif
/* 31 bit s390 suupport is not yet included, it may never be.
#ifdef __s390__
long long _div64 (long long u, long long v);
long long _mod64 (long long u, long long v);
#endif
*/
/** Divide x by y. In case of division-by-zero,
* set context error string, and return 0
*/
int64_t _stp_div64 (const char **error, int64_t x, int64_t y)
{
// check for division-by-zero
if (unlikely (y == 0)) {
if (error) *error = "division by 0";
return 0;
}
if (unlikely (y == -1))
return -x;
#ifdef __LP64__
return x/y;
#else
if (likely ((x >= LONG_MIN && x <= LONG_MAX) &&
(y >= LONG_MIN && y <= LONG_MAX))) {
return (long)x / (long)y;
} else
return _div64 (x, y);
#endif
}
/** Modulo x by y. In case of division-by-zero,
* set context error string, and return any 0
*/
int64_t _stp_mod64 (const char **error, int64_t x, int64_t y)
{
// check for division-by-zero
if (unlikely (y == 0)) {
if (error) *error = "division by 0";
return 0;
}
if (unlikely (y == 1 || y == -1))
return 0;
#ifdef __LP64__
return x%y;
#else
if (likely ((x >= LONG_MIN && x <= LONG_MAX) &&
(y >= LONG_MIN && y <= LONG_MAX))) {
return (long)x % (long)y;
} else
return _mod64 (x, y);
#endif
}
#ifndef _STP_TEST_
/** Return a random integer between -n and n.
* @param n how far from zero to go. Make it positive but less than a million or so.
*/
int _stp_random_pm (int n)
{
static unsigned long seed;
static int initialized_p = 0;
if (unlikely (! initialized_p)) {
seed = (unsigned long) jiffies;
initialized_p = 1;
}
/* from glibc rand man page */
seed = seed * 1103515245 + 12345;
return (seed % (2*n+1)-n);
}
#endif /* _STP_TEST_ */
#ifdef __i386__
/* 64-bit division functions extracted from libgcc */
typedef long long DWtype;
typedef unsigned long long UDWtype;
typedef unsigned long UWtype;
typedef long Wtype;
typedef unsigned int USItype;
#ifdef _BIG_ENDIAN
struct DWstruct {Wtype high, low;};
#else
struct DWstruct {Wtype low, high;};
#endif
#define W_TYPE_SIZE 32
typedef union
{
struct DWstruct s;
DWtype ll;
} DWunion;
/* these are the i386 versions of these macros from gcc/longlong.h */
#define umul_ppmm(w1, w0, u, v) \
__asm__ ("mull %3" \
: "=a" ((USItype) (w0)), \
"=d" ((USItype) (w1)) \
: "%0" ((USItype) (u)), \
"rm" ((USItype) (v)))
#define sub_ddmmss(sh, sl, ah, al, bh, bl) \
__asm__ ("subl %5,%1\n\tsbbl %3,%0" \
: "=r" ((USItype) (sh)), \
"=&r" ((USItype) (sl)) \
: "0" ((USItype) (ah)), \
"g" ((USItype) (bh)), \
"1" ((USItype) (al)), \
"g" ((USItype) (bl)))
#define udiv_qrnnd(q, r, n1, n0, dv) \
__asm__ ("divl %4" \
: "=a" ((USItype) (q)), \
"=d" ((USItype) (r)) \
: "0" ((USItype) (n0)), \
"1" ((USItype) (n1)), \
"rm" ((USItype) (dv)))
#define count_leading_zeros(count, x) \
do { \
USItype __cbtmp; \
__asm__ ("bsrl %1,%0" \
: "=r" (__cbtmp) : "rm" ((USItype) (x))); \
(count) = __cbtmp ^ 31; \
} while (0)
inline UDWtype _stp_udivmoddi4 (UDWtype n, UDWtype d, UDWtype *rp)
{
const DWunion nn = {.ll = n};
const DWunion dd = {.ll = d};
DWunion ww,rr;
UWtype d0, d1, n0, n1, n2;
UWtype q0, q1;
UWtype b, bm;
d0 = dd.s.low;
d1 = dd.s.high;
n0 = nn.s.low;
n1 = nn.s.high;
if (d1 == 0) {
if (d0 > n1) {
/* 0q = nn / 0D */
udiv_qrnnd (q0, n0, n1, n0, d0);
q1 = 0;
/* Remainder in n0. */
} else {
/* qq = NN / 0d */
if (d0 == 0)
d0 = 1 / d0; /* Divide intentionally by zero. */
udiv_qrnnd (q1, n1, 0, n1, d0);
udiv_qrnnd (q0, n0, n1, n0, d0);
/* Remainder in n0. */
}
if (rp != 0) {
rr.s.low = n0;
rr.s.high = 0;
*rp = rr.ll;
}
} else {
if (d1 > n1) {
/* 00 = nn / DD */
q0 = 0;
q1 = 0;
/* Remainder in n1n0. */
if (rp != 0) {
rr.s.low = n0;
rr.s.high = n1;
*rp = rr.ll;
}
} else {
/* 0q = NN / dd */
count_leading_zeros (bm, d1);
if (bm == 0) {
/* From (n1 >= d1) /\ (the most significant bit of d1 is set),
conclude (the most significant bit of n1 is set) /\ (the
quotient digit q0 = 0 or 1).
This special case is necessary, not an optimization. */
/* The condition on the next line takes advantage of that
n1 >= d1 (true due to program flow). */
if (n1 > d1 || n0 >= d0) {
q0 = 1;
sub_ddmmss (n1, n0, n1, n0, d1, d0);
} else
q0 = 0;
q1 = 0;
if (rp != 0) {
rr.s.low = n0;
rr.s.high = n1;
*rp = rr.ll;
}
} else {
UWtype m1, m0;
/* Normalize. */
b = W_TYPE_SIZE - bm;
d1 = (d1 << bm) | (d0 >> b);
d0 = d0 << bm;
n2 = n1 >> b;
n1 = (n1 << bm) | (n0 >> b);
n0 = n0 << bm;
udiv_qrnnd (q0, n1, n2, n1, d1);
umul_ppmm (m1, m0, q0, d0);
if (m1 > n1 || (m1 == n1 && m0 > n0)) {
q0--;
sub_ddmmss (m1, m0, m1, m0, d1, d0);
}
q1 = 0;
/* Remainder in (n1n0 - m1m0) >> bm. */
if (rp != 0) {
sub_ddmmss (n1, n0, n1, n0, m1, m0);
rr.s.low = (n1 << b) | (n0 >> bm);
rr.s.high = n1 >> bm;
*rp = rr.ll;
}
}
}
}
ww.s.low = q0; ww.s.high = q1;
return ww.ll;
}
long long _div64 (long long u, long long v)
{
long c = 0;
DWunion uu = {.ll = u};
DWunion vv = {.ll = v};
DWtype w;
if (uu.s.high < 0)
c = ~c,
uu.ll = -uu.ll;
if (vv.s.high < 0)
c = ~c,
vv.ll = -vv.ll;
w = _stp_udivmoddi4 (uu.ll, vv.ll, (UDWtype *) 0);
if (c)
w = -w;
return w;
}
long long _mod64 (long long u, long long v)
{
long c = 0;
DWunion uu = {.ll = u};
DWunion vv = {.ll = v};
DWtype w;
if (uu.s.high < 0)
c = ~c,
uu.ll = -uu.ll;
if (vv.s.high < 0)
vv.ll = -vv.ll;
(void) _stp_udivmoddi4 (uu.ll, vv.ll, (UDWtype*)&w);
if (c)
w = -w;
return w;
}
#endif /* __i386__ */
#endif /* _ARITH_C_ */
|
