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
|
#include <stdio.h>
#include <math.h>
#include <asm/msr.h>
#include "resample.h"
#define CLK 300.e6
#define LOOPS 20000
typedef double ( *ddf ) ( double );
float a1 [256];
float a2 [256];
void init ( void )
{
int i;
for ( i = 0; i < sizeof(a1)/sizeof(*a1); i++ ) {
a1 [i] = sin(i)+0.2*sin(1.8*i)+log(2+i);
a2 [i] = cos(0.1*i);
}
}
void test ( int no, scalar_t f )
{
unsigned long long t1;
unsigned long long t2;
unsigned long long t3;
unsigned long long t4;
int l;
double last = 0;
double curr = 0;
printf ( "[%3u] %22.14f\t\t", no, (double)f (a1,a2) );
fflush ( stdout );
do {
rdtscll (t1);
l = LOOPS;
do
;
while (--l);
rdtscll (t2);
rdtscll (t3);
l = LOOPS;
do
f(a1,a2), f(a1,a2), f(a1,a2), f(a1,a2);
while (--l);
rdtscll (t4);
last = curr;
curr = (t4-t3-t2+t1) / CLK / LOOPS / 4 * 1.e9;
} while ( fabs(curr-last) > 1.e-4 * (curr+last) );
printf ("%8.2f ns\n", (curr+last) / 2 );
}
void testn ( scalarn_t f )
{
unsigned long long t1;
unsigned long long t2;
unsigned long long t3;
unsigned long long t4;
int l;
int i;
double last = 0;
double curr = 0;
for ( i = 1; i <= 64; i += i<6 ? 1 : i<8 ? 2 : i ) {
printf ( "[%3u] %22.14f\t\t", 4*i, (double)f (a1,a2,i) );
fflush ( stdout );
do {
rdtscll (t1);
l = LOOPS;
do
;
while (--l);
rdtscll (t2);
rdtscll (t3);
l = LOOPS;
do
f(a1,a2,i), f(a1,a2,i), f(a1,a2,i), f(a1,a2,i);
while (--l);
rdtscll (t4);
last = curr;
curr = (t4-t3-t2+t1) / CLK / LOOPS / 4 * 1.e9;
} while ( fabs(curr-last) > 1.e-4 * (curr+last) );
printf ("%8.2f ns\n", (curr+last) / 2 );
}
}
void test2 ( const char* name, ddf f )
{
int i;
double x;
printf ( "\n%%%% %s\n\n", name );
for ( i = -1000; i <= 1000; i++ ) {
x = 1.e-3 * i;
printf ( "%5d\t%12.8f\t%12.8f\t%12.8f\n", i, f(x), (f(x+5.e-5) - f(x-5.e-5))*1.e+4, (f(x+1.e-4) + f(x-1.e-4) - 2*f(x))*5.e+7 );
}
printf ( "%%%%\n" );
fflush ( stdout );
}
int main ( int argc, char** argv )
{
#if 0
test2 ( "Hann", hanning );
test2 ( "Hamm", hamming );
test2 ( "BM", blackman );
test2 ( "BM1",blackman1 );
test2 ( "BM2",blackman2 );
test2 ( "BMH N",blackmanharris_nuttall );
test2 ( "MNH Min",blackmanharris_min4 );
#else
init ();
test ( 4, scalar04_float32 );
test ( 4, scalar04_float32_i387 );
test ( 4, scalar04_float32_3DNow );
test ( 4, scalar04_float32_SIMD );
test ( 8, scalar08_float32 );
test ( 8, scalar08_float32_i387 );
test ( 8, scalar08_float32_3DNow );
test ( 8, scalar08_float32_SIMD );
test ( 12, scalar12_float32 );
test ( 12, scalar12_float32_i387 );
test ( 12, scalar12_float32_3DNow );
test ( 12, scalar12_float32_SIMD );
test ( 16, scalar16_float32 );
test ( 16, scalar16_float32_i387 );
test ( 16, scalar16_float32_3DNow );
test ( 16, scalar16_float32_SIMD );
test ( 20, scalar20_float32 );
test ( 20, scalar20_float32_i387 );
test ( 20, scalar20_float32_3DNow );
test ( 20, scalar20_float32_SIMD );
test ( 24, scalar24_float32 );
test ( 24, scalar24_float32_i387 );
test ( 24, scalar24_float32_3DNow );
test ( 24, scalar24_float32_SIMD );
testn( scalar4n_float32 );
testn( scalar4n_float32_i387 );
testn( scalar4n_float32_3DNow );
testn( scalar4n_float32_SIMD );
#endif
return 0;
}
/* end of scalartest.c */
|
