[Bug fortran/30528] New: INTEGER(2) overflow

[Bug fortran/30528] New: INTEGER(2) overflow

[remi dot mommsen at gmx dot net]

These lines cannot be compiled with gfortran (GNU Fortran 95 (GCC) 4.2.0
20070117 (prerelease)) on a Mac OS X 10.4 (powerpc): 

      INTEGER*2 IWD1
      IWD1 = 32768
      END

It fails with

      IWD1 = 32768
            1
Error: Arithmetic overflow converting INTEGER(4) to INTEGER(2) at (1)


This works for g77 (GNU Fortran (GCC) 3.4.3) as well as with gfortran (and g77)
on Mac OS X 10.4 on intel (i686-apple-darwin8).

All fortran compilers have been compiled within the Fink framework.


-- 
           Summary: INTEGER(2) overflow
           Product: gcc
           Version: 4.2.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: fortran
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: remi dot mommsen at gmx dot net
 GCC build triplet: powerpc-apple-darwin8
  GCC host triplet: powerpc-apple-darwin8
GCC target triplet: powerpc-apple-darwin8


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30528

[Bug fortran/30528] INTEGER(2) overflow

[jvdelisle at gcc dot gnu dot org]

------- Comment #1 from jvdelisle at gcc dot gnu dot org  2007-01-21 20:11 -------
The largest positive integer that can be represented by that kind type is
32767, so gfortran is correctly reporting an error.

Use -fno-range-check to bypass this.


-- 

jvdelisle at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |WAITING


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30528

[Bug fortran/30528] INTEGER(2) overflow

[fxcoudert at gcc dot gnu dot org]

------- Comment #2 from fxcoudert at gcc dot gnu dot org  2007-01-21 20:14 -------
First, it should be noted that it's not legal code. Indeed, you might (or not)
be surprised by the behaviour of the code in question:

pito /tmp $ cat a.f 
      INTEGER*2 IWD1
      IWD1 = 32768
      PRINT *, IWD1
      END
pito /tmp $ g77 a.f && ./a.out
 -32768

This check is new in gfortran, the behaviour is intentional but can be
specificaly disabled using -fno-range-check:

$ gfortran -fno-range-check a.f && ./a.out
 -32768


Closing bug report as INVALID. Making note to document this in the release
notes.


-- 

fxcoudert at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|WAITING                     |RESOLVED
  GCC build triplet|powerpc-apple-darwin8       |
   GCC host triplet|powerpc-apple-darwin8       |
 GCC target triplet|powerpc-apple-darwin8       |
         Resolution|                            |INVALID


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30528