https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69101
--- Comment #2 from kargl at gcc dot gnu.org --- (In reply to kargl from comment #1) > This has nothing to do with either -fdefault-integer-8 or > -finteger-4-integer-8. The IEEE module is not implemented > correctly. F2003: 14.10 Specifications of the procedures In the detailed descriptions below, procedure names are generic and are not specific. On x86_64, there are 5 integer kind type parameters. For IEEE_SELECTED_REAL_KIND(P,R), gfortran either needs 25 helper functions to accommodate the 25 combinations of P and R or a helper conversion function that forces R and P to a specific integer kind type.