[Bug libgcc/105669] DFP to HF (_Float16) conversions use incorrect rounding

[jsm28 at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105669

Joseph S. Myers <jsm28 at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Last reconfirmed|                            |2022-05-20
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |NEW

--- Comment #1 from Joseph S. Myers <jsm28 at gcc dot gnu.org> ---
Example test illustrating this bug, at least on x86_64 (the DFP input is very
close to half way between two consecutive _Float16 values):

extern void exit (int);
extern void abort (void);
#define DCONST 7.340312e-5DF
static const _Float16 f1 = DCONST;
volatile _Decimal32 d = DCONST;
int
main (void)
{
  volatile _Float16 f2;
  f2 = d;
  if (f1 == f2)
    exit (0);
  else
    abort ();
}