https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108680
--- Comment #9 from Jerry DeLisle <jvdelisle at gcc dot gnu.org> --- We can have only one default integer otherwise its not a default. Our default integer is KIND=4 The RANGE of KIND=4 integer is 9, so we exceed the requirement for at least a decimal range of 5. RANGE is defined as INT (LOG10 (HUGE (X))). I am thinking that we should just document that -fdefault-integer-8 is incompatible with IO statements. Frankly, using it is improper programming anyway as the language provides methods intended to allow users to change the working kinds using a user defined kind parameter. We probably should not have this option because it is problematic. We could throw an error if a user uses -fdefault-integer-8 with I/O statements with usefule message.
