https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122325
Richard Biener <rguenth at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Assignee|unassigned at gcc dot gnu.org |rguenth at gcc dot
gnu.org
Status|UNCONFIRMED |ASSIGNED
Last reconfirmed| |2025-10-20
Ever confirmed|0 |1
--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
Confirmed. So with -O -ffast-math we first handle
(gdb) p *decoded
$3 = {opt_index = 1317, warn_message = 0x0, arg = 0x0,
orig_option_with_args_text = 0x4ee4410 "-fno-fast-math", canonical_option = {
0x4ee4410 "-fno-fast-math", 0x0, 0x0, 0x0},
canonical_option_num_elements = 1, value = 0, mask = 0, errors = 4}
then
(gdb) p *decoded
$4 = {opt_index = 1317, warn_message = 0x0, arg = 0x0,
orig_option_with_args_text = 0x4ee3da0 "-ffast-math", canonical_option = {
0x467bb42 "-ffast-math", 0x0, 0x0, 0x0},
canonical_option_num_elements = 1, value = 1, mask = 0, errors = 0}
which correctly sets opts->x_flag_complex_method = 0;
But then finish_options comes along:
/* Use a frontend provided default for the complex eval method. */
if (!opts_set->x_flag_complex_method)
opts->x_flag_complex_method = opts->x_flag_default_complex_method;
Previously we had at this point
/* With -fcx-limited-range, we do cheap and quick complex arithmetic. */
if (opts->x_flag_cx_limited_range)
opts->x_flag_complex_method = 0;
else if (opts_set->x_flag_cx_limited_range)
opts->x_flag_complex_method = opts->x_flag_default_complex_method;
/* With -fcx-fortran-rules, we do something in-between cheap and C99. */
if (opts->x_flag_cx_fortran_rules)
opts->x_flag_complex_method = 1;
else if (opts_set->x_flag_cx_fortran_rules)
opts->x_flag_complex_method = opts->x_flag_default_complex_method;
IMO flag_default_complex_method should go away.
