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
|
/*
** float-rounder-bug.c
**
** Martin Brain
** martin.brain@cs.ox.ac.uk
** 20/05/13
**
** Another manifestation of the casting bug.
** If the number is in (0,0x1p-126) it is rounded to zero rather than a subnormal number.
*/
#define FULP 1
void bug (float min) {
__CPROVER_assume(min == 0x1.fffffep-105f);
float modifier = (0x1.0p-23 * (1<<FULP));
float ulpdiff = min * modifier;
assert(ulpdiff == 0x1p-126f); // Should be true
return;
}
void bugBrokenOut (float min) {
__CPROVER_assume(min == 0x1.fffffep-105f);
float modifier = (0x1.0p-23 * (1<<FULP));
double dulpdiff = (double)min * (double)modifier; // Fine up to here
float ulpdiff = (float)dulpdiff; // Error
assert(ulpdiff == 0x1p-126f); // Should be true
return;
}
void bugCasting (double d) {
__CPROVER_assume(d == 0x1.fffffep-127);
float f = (float) d;
assert(f == 0x1p-126f); // Should be true
return;
}
int main (void) {
float f;
bug(f);
float g;
bugBrokenOut(g);
double d;
bugCasting(d);
return 1;
}
|
