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
|
/* { dg-do run } */
/* { dg-require-effective-target sse4 } */
/* { dg-options "-O2 -msse4.1" } */
/* { dg-skip-if "no M_PI" { vxworks_kernel } } */
#include "sse4_1-check.h"
#include <smmintrin.h>
#include <math.h>
#include <string.h>
#define NUM 64
static void
init_round (float *src)
{
int i, sign = 1;
float f = rand ();
for (i = 0; i < NUM; i++)
{
src[i] = (i + 1)* f * M_PI * sign;
if (i < (NUM / 2))
{
if ((i % 6) == 0)
f = f * src[i];
}
else if (i == (NUM / 2))
f = rand ();
else if ((i % 6) == 0)
f = 1 / (f * (i + 1) * src[i] * M_PI *sign);
sign = -sign;
}
}
static float
do_round (float f, int type)
{
short saved_cw, new_cw, clr_mask;
float ret;
if ((type & 4))
{
type = 0;
clr_mask = 0xFFFF;
}
else
{
type = 0x003F | ((type & 3) << 10);
clr_mask = ~0x0C3F;
}
__asm__ ("flds %0" : : "m" (*&f));
__asm__ ("fstcw %0" : "=m" (*&saved_cw));
new_cw = saved_cw & clr_mask;
new_cw |= type;
__asm__ ("fldcw %0" : : "m" (*&new_cw));
__asm__ ("frndint\n"
"fstps %0\n" : "=m" (*&ret));
__asm__ ("fldcw %0" : : "m" (*&saved_cw));
return ret;
}
static void
sse4_1_test (void)
{
int i, j;
float f;
union
{
__m128 x[NUM / 4];
float f[NUM];
} dst, src;
init_round (src.f);
memset (&dst, 0, NUM * sizeof(float));
for (i = 0; i < NUM / 4 ; i++)
dst.x[i] = _mm_round_ss (dst.x[i], src.x[i], _MM_FROUND_RINT);
for (i = 0; i < NUM; i += 4)
{
for (j = 0; j < 3; j++)
if (dst.f[i + j + 1] != 0.0)
abort ();
f = do_round (src.f[i], 0x04);
if (f != dst.f[i])
abort ();
}
for (i = 0; i < NUM / 4 ; i++)
dst.x[i] = _mm_round_ss (dst.x[i], src.x[i], _MM_FROUND_NEARBYINT);
for (i = 0; i < NUM; i += 4)
{
for (j = 0; j < 3; j++)
if (dst.f[i + j + 1] != 0.0)
abort ();
f = do_round (src.f[i], 0x0c);
if (f != dst.f[i])
abort ();
}
}
|
