[Bug fortran/34505] New: -std=f*: FLOAT accepts non-integers and SNGL non-double precision reals

[burnus at gcc dot gnu dot org]

Per "13.6 Specific names for standard intrinsic functions" (Fortran 2003)

FLOAT is a specific function of REAL which takes as argument
      only a default-kind integer.
SNGL  is a specific function of REAL which takes as argument
      only a double precision real.

Current implementation is that it takes any kind of int/real/complex as REAL
does.


-- 
           Summary: -std=f*:  FLOAT accepts non-integers and SNGL non-double
                    precision reals
           Product: gcc
           Version: 4.3.0
            Status: UNCONFIRMED
          Keywords: accepts-invalid, diagnostic
          Severity: normal
          Priority: P3
         Component: fortran
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: burnus at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34505