https://gcc.gnu.org/bugzilla/show_bug.cgi?id=117805
kargls at comcast dot net changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |kargls at comcast dot net
--- Comment #6 from kargls at comcast dot net ---
(In reply to mjr19 from comment #0)
> Recently the Fortran Standards Committee, J3, has clarified the application
> of IEEE-754 to the complex type in Fortran.
Technically, J3 declined to offer an interpretation. Not to mention
that the logic offered in J3/24-179 is flawed.
Of course, IEEE-754 will not considered complex numbers. It deals
with the mapping of real numbers to a finite set of floating point
numbers. For J3 to write
There is no question that the mathematical integers, real numbers, and
complex numbers, do not include the IEEE special values of infinity or
NaN.
is simply silly. The real numbers do include an infinity. IEEE-754
provides a mapping of infinity to a collections of bits. In fact,
IEEE-754 2008 states
The mathematical structure underpinning the arithmetic in this standard is
the extended reals, that is, the set of real numbers together with positive
and negative infinity.
The foundations of mathematics most certainly has the concepts
of +-0, +-inf, and even NaN (i.e., an indeterminate form, see
L'Hopital's rule).
Use -Ofast pr --fast-math if you want a possibly incorrect result
fast.
