[FPLO] compiling FPLO5.00-20 on a 64 bit system

Klaus Koepernik K.Koepernik at ifw-dresden.de
Tue Nov 18 10:18:15 CET 2008


I didn't follow the discussion carefully
just in case
Some segfaults are stack-size restrictions + compiler stupidity

solution: compile with  
	-heap-arrays 100

this will put all locacl variables on heap and thus keep the
stack small.

Klaus



On Mon, Nov 17, 2008 at 07:38:38PM +0100, Neculai Plugaru wrote:
> HI Michael
> 
> Thank you for the message, and also Thanks to Sushil ! 
>  
> Yes, I already did it, and now I struggle with the segmentation fault 
> problem ! Ok, I will follow the workaround indicated by Ulli to solve this 
> problem. 
> I have also checked that fplo7 compiled with ifort is much faster than when 
> compiled using the open source compilers.  
> 
> Best regards !
> Nucu Plugaru
> 
> 
> On Monday 17 November 2008 16:33, Dr. Michael Frotscher wrote:
> 
> > On Tuesday November 11 2008 n.plugaru wrote:
> > 
> > > GNU Fortran 95 (GCC) 4.1.2 20061115 (prerelease) (Debian 4.1.1-21)
> > 
> > Go and get the intel ifc fortran compiler (I think, it's now called ifort). 
> > It's free for academic institutions and works well.
> > 
> > Sincerely,
> > -- 
> > Dr. Michael Frotscher
> > Eduard-Zintl-Institute for Inorganic and Physical Chemistry
> > Darmstadt University of Technology, Germany
> > 
> > To unsubscribe from list fplo-users please send mail to 
> > fplo-users-subscribe at mailbox.ifw-dresden.de
> > with "unsubscribe fplo-users" in the body.
> > 
> 
> To unsubscribe from list fplo-users please send mail to 
> fplo-users-subscribe at mailbox.ifw-dresden.de
> with "unsubscribe fplo-users" in the body.

-- 

----------------------------------------------------
Dr. Klaus Koepernik
Inst. for Theoretical Solid State Physics
Leibniz Institute for Solid State and 
Materials Research Dresden
P.O. Box 270016
D-01171 Dresden, Germany
email: k.koepernik at ifw-dresden.de
Tel:   Germany (0)351 4659360


More information about the fplo-users mailing list