https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93727
--- Comment #8 from Jerry DeLisle <jvdelisle at gcc dot gnu.org> ---
I have had to explore this a bit to get to a point of understanding. The
following C example shows building the HEX float strings and the incantations
needed. For those systems that support the 17 byte float format, I have not
found what is needed yet.
#include <stdio.h>
#include <quadmath.h>
int main()
{
char buf[128];
float x = 1.0f / 3.0f;
double y = 1.0l / 3.0l;
long double z = 1.0L / 3.0L;
__float128 q = 1.0Q / 3.0Q;
printf(" FLOAT: %.7A\n", x);
printf(" DOUBLE: %.15lA\n", y);
printf("LONG DBL: %.18LA\n", z);
int n = quadmath_snprintf (buf, sizeof buf, "%.30QA", q);
if ((size_t) n < sizeof buf)
printf ("QUAD DBL: %s\n", buf);
printf(" 1234567890123456789012345678901234567890\n");
printf(" FLOAT: %.9f\n", x);
printf(" DOUBLE: %.18lf\n", y);
printf("LONG DBL: %.22Lf\n", z);
printf("QUAD DBL: %.36Qf \n", q);
}
I still need to look at the code for the kind 17 and where to test it.
$ gcc hexfloat.c -lquadmath
$ ./a.out
FLOAT: 0X1.5555560P-2
DOUBLE: 0X1.555555555555500P-2
LONG DBL: 0XA.AAAAAAAAAAAAAAB000P-5
QUAD DBL: 0X1.555555555555555555555555555500P-2
1234567890123456789012345678901234567890
FLOAT: 0.333333343
DOUBLE: 0.333333333333333315
LONG DBL: 0.3333333333333333333424
QUAD DBL: 0.333333333333333333333333333333333317
