[CPMD-list] about debugging in cpmd

Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu
Sat Jun 30 06:04:09 CEST 2007


On Fri, 29 Jun 2007, Bin Pan wrote:

BP> Dear CPMD users,

dear bin,

BP> I am currently working on modifying some code in CPMD. Unfortunately 
BP> I haven't found a good way in debugging CPMD.
BP> I compiled CPMD using PGI fortran compiler so that I have to use its 
BP> debugger pgdbg. What I found strange is that I can not set
BP> breakpoints at anywhere I want them to be. When I pressed the button 
BP> "next", meaning only to run the next single line of code, the program seemed
BP> to jump several lines and stopped at random a new position.

did you disable optimization when compiling CPMD for debugging?
generally, i would recommend to build CPMD with gfortran/g95 
and/or debug with gdb or its graphical frontends. 

BP> How can I make it only run a single line of code, just like in GNU 
BP> fortran debugger gdb, when typing "n", it only runs a single step?
BP> Did anybody have similar experience in having this problem?

i find the PGI tools inferior for debugging than the GNU tools,
so i don't use them. gfortran and g95 also have a large variety
in warning and code quality checks, that can be extremely helpful
in debugging code. i only use PGI when there is not viable alternative.

BP> I would appreciate to know how you guys debug your code using PGI 
BP> compiler and debugger.

i suspect that nobody does.

cheers,
   axel.

BP> Thanks a lot!
BP> 
BP> Best regards,
BP> Bin.
BP> 
BP> _______________________________________________
BP> CPMD-list mailing list
BP> CPMD-list at cpmd.org
BP> http://cpmd.org/mailman/listinfo/cpmd-list
BP> 

-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.



More information about the CPMD-list mailing list