| 3:13:00 PM PDT - Wed, Sep 21st 2016 |
|
Ben:
A few years ago, I worked around these issues and compiled NWChem with Cray Fortran. The resulting binary gave wrong answers for the DFT code. Thus, supporting Cray Fortran is not as simple as just resolving recursive preprocessing (which, by the way, can be solved the same way NWChem works around limitations in the IBM XLF preprocessor).
I mention this in hopes of saving someone else the trouble of repeating my leap from the frying pan into the fire.
|