[Bug fortran/29561] New: hexadecimal constant problem

["kloedej at knmi dot nl" <gcc-bugzilla at gcc dot gnu dot org>]

The following sample code:

program testhexconstant

  integer, parameter :: i4_ = Selected_Int_Kind( 9)  

  ! works fine
  integer(i4_), parameter :: pos_nan_lower = z"7F800001"
  ! triggers an error
  integer(i4_), parameter :: pos_nan_lower = z"FF800001"

end program testhexconstant

triggers this error:
gfortran -o testhex TestHex.F90
 In file TestHex.F90:8

  integer(i4_), parameter :: b = z"FF800001"
                               1
Error: Arithmetic overflow converting INTEGER(8) to INTEGER(4) at (1)

for the gfortran version downloaded today (20061023) from 
http://quatramaran.ens.fr/~coudert/gfortran/gfortran-linux.tar.gz

while the previous version that I had installed (dated 20060914) downloaded
from the same place, just works fine with this code.
This hex value should fit nicely anyway in a 4 byte integer, so I don't really
understand what is happening here.

regards,

Jos de Kloe


-- 
           Summary: hexadecimal constant problem
           Product: gcc
           Version: 4.2.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: fortran
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: kloedej at knmi dot nl


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29561