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
|
/* ************************************************************************
* Copyright 2013 Advanced Micro Devices, Inc.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
* ************************************************************************/
#ifndef CLBLAS_STDDEF_H_
#define CLBLAS_STDDEF_H_
static __inline size_t
szmin(size_t a, size_t b)
{
return (a <= b ? a : b);
}
static __inline size_t
szmax(size_t a, size_t b)
{
return (a >= b ? a : b);
}
static __inline unsigned int
umin(unsigned int a, unsigned int b)
{
return (a <= b ? a : b);
}
static __inline unsigned int
umax(unsigned int a, unsigned int b)
{
return (a >= b ? a : b);
}
static __inline void
uswap(unsigned int *a, unsigned int *b)
{
unsigned int tmp;
tmp = *a;
*a = *b;
*b = tmp;
}
static __inline size_t
roundDown(size_t a, size_t b)
{
return (a / b * b);
}
static __inline size_t
roundUp(size_t a, size_t b)
{
return (a + b - 1) / b * b;
}
static __inline size_t
divRoundUp(size_t a, size_t b)
{
return (a / b) + (a % b != 0);
}
static __inline int
isRoundedPow2(size_t a)
{
return ((a & (a - 1)) == 0);
}
/*
* Return zero based sequential number of the highest set bit the
* number. If the number is 0, then the function returns -1.
*/
static __inline int
findHighestSetBit(size_t a)
{
int n = (sizeof(size_t) * 8 - 1);
size_t s = (size_t)1 << n;
for (; (s != 0) && !(s & a); s >>= 1) {
n--;
}
return (s == 0) ? -1 : n;
}
static __inline size_t
roundDownPow2(size_t a)
{
size_t s;
if (isRoundedPow2(a)) {
return a;
}
s = (size_t)1 << (sizeof(size_t) * 8 - 1);
// find the highest non zero bit
for (; (s != 0) && !(s & a); s >>= 1);
return s;
}
/*
* With BLAS we never deal with so large number sufficient for overflowing.
* So, it's safe
*/
static __inline size_t
roundUpPow2(size_t a)
{
size_t s;
if (isRoundedPow2(a)) {
return a;
}
s = roundDownPow2(a);
return (s << 1);
}
#endif /* CLBLAS_STDDEF_H_ */
|
