[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: gfortran.

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: Re: gfortran.
  • Date: Thu, 23 Aug 2007 10:28:38 -0600
Stonie,

I was able to run GDDIAG here using your functions below, and we pounded
on it pretty hard during the workshop. I ran the binary I version as
posted 8/21/07 though valgrind but nothing popped out. I'll take a look
at the code and see if its just a latent bug unrelated to g77/gfortran.

Steve



On Thu, 2007-08-23 at 11:13 -0500, Stonie Cooper wrote:
> Steve,
> 
> I spoke too soon.
> 
> This is on a RHEL clone (CentOS v4.5), with gcc version 4.1.1  
> 20070105 (Red Hat 4.1.1-53).
> 
> I compiled with gfortran.
> 
> gddiag seg faults on computations (replicates every time):
> 
> gddiag << end_gddiag
> GDFILE   = /data/ldm/gempak/model/ruc2236/2007082312_ruc2236.gem
> GDOUTF   = temp.gem
> GFUNC    = sub(add(sub(TMPC@850,TMPC@500),DWPC@850),sub 
> (TMPC@700,DWPC@700))
> GDATTIM  = 070823/1200F001
> GLEVEL   = 0
> GVCORD   = PRES
> GRDNAM   = KINDX@0%NONE^070823/1200F001
> GRDTYP   =
> GPACK    =
> GRDHDR   =
> PROJ     =
> GRDAREA  =
> KXKY     =
> MAXGRD   =
> CPYFIL   =
> ANLYSS   =
> RUN
> 
>   19823 Segmentation fault      gddiag  << end_gddiag
> 
> 
> I took this system off-line, recompiled with g77, and the seg faults  
> went away.  FYI.  The g77 was 3.4.6.
> 
> Incidently, this same "feature" does NOT appear on the Gentoo system,  
> compiled with gfortran 4.1.2.  As a whole, we tend to have a lot less  
> problems on Gentoo than the other distributions.  This particular  
> problem has only appeared on the RHEL clone, thus far.
> 
> Stonie
-- 
Steve Chiswell <address@hidden>
Unidata