[Wien] Problems with Wien2k_06.4

Peter Blaha pblaha at theochem.tuwien.ac.at
Thu Sep 7 16:02:11 CEST 2006


If I understand you correctly, your compiler crashes. In which subroutine 
does it happen ?

> the code in SRC_ncmsymmetry compiles on my computer (D820, Wien2k_06.4, SUSE 10.1, ifc 9.0.031, mkl 8.0 .006) only correctly if I use -O0, that is without any optimizations. In my case I changed just that makefile and kept all others
> with full optimization.
> Is there anything in the code changed that might lead to overoptimization with -O2 (default) or -O3, 
> sometimes the compiler does not like how continuations in
> lines with subroutine calls are made
>             call xyz ( a, b,
>      &                 c, d) 
> should be fine, whereas sometimes (I could not figure out when)
>             call xyz 
>      &      ( a, b, c, d)
> or 
>             call xyz ( 
>      &                 a, b, c, d)
> leads to strange errors or compiler crashes (the last time I had this was with ifc 8.0,
> but may be its still there just in a different way). 

language setting influences only the scripts which use "awk", which is not
the case in optimize.job.
I seem to remember that in some old version there were 2 \\ characters as
continuation characters ?

> The strange gcc-lib error that I experienced (see old mail) appeared
> when I used the script file for Volume optimization produced by an older Version of this year (sorry its already deleted, I cannot figure out which one). After I produced the script file with the new Version, it was running fine, without errors.
> Is there anything changed in the code that may conflict with language settings?


                                      P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-15671             FAX: +43-1-58801-15698
Email: blaha at theochem.tuwien.ac.at    WWW: http://info.tuwien.ac.at/theochem/
--------------------------------------------------------------------------


More information about the Wien mailing list