This is the mail archive of the
fortran@gcc.gnu.org
mailing list for the GNU Fortran project.
Re: BUG in gfortran4.1.1/intrinsic rrspacing
- From: Steve Kargl <sgk at troutmask dot apl dot washington dot edu>
- To: Matthias Langer <matthias dot langer at zamg dot ac dot at>
- Cc: fortran at gcc dot gnu dot org
- Date: Tue, 20 Jun 2006 09:32:47 -0700
- Subject: Re: BUG in gfortran4.1.1/intrinsic rrspacing
- References: <1150811250.6948.1.camel@zaupc10.zamg.ac.at>
On Tue, Jun 20, 2006 at 03:47:29PM +0200, Matthias Langer wrote:
> ! ...test the F95-intrinsic function "rrspacing"
> ! gfortran produces an internal compiler error, if rrspacing is used
> with
> ! a constant or a parameter as argument.
> ! >gfortran rrspacing_test.f90
> ! >rrspacing_test.f90:0: internal compiler error: Segmentation fault
> !ScientificLinux-4.3 on Pentium4/ 2.66GHz PC
> !Linux 2.6.9-34.0.1.EL #1 Wed May 24 13:59:58 CDT 2006 i686 i686 i386
> GNU/Linux
> !GNU Fortran 95 (GCC) 4.1.1/release 24may2006
> program RRSPACING_test
> implicit none
> real, parameter :: xpar=1.0 ! parameter/constant
> real :: xvar
> real :: ydum
> !=>
> xvar = 1.0
> ydum = rrspacing(xvar) ! works well with variables
> !*** doesn't work with parameters or
> constants**********************************
> ! ydum = rrspacing(xpar) ! parameter
> ! ydum = rrspacing(1.0) ! constant expression
> print *, ydum
> END program RRSPACING_test
> !
Looks like a bug. Can you file a buzilla report?
--
Steve