OpenRSP/HODI: Memory issue for geo. properties?
@kdu002 @simensr There was a problem that I think you Karen found about 3 months ago where there was a traceback leading to some memory/segfault problem in HODI for calculation of high-order GEO properties. As a result, I implemented a disabling of high-order GEO stuff from OpenRSP until this was looked at.
- I can't find an issue about this so I'm making one now.
- I can't find the original messages about the error in my archives, so Karen or Simen, do you remember this problem and do you have a backtrace (and other calculation background information from Karen) to post here for reference?
- I made a mistake when I disabled high GEO orders and in fact was too lenient, so that the present code does in fact allow one higher GEO order than intended.
Ideally we would take care of this problem before the release and that probably means Simen looking at the relevant parts of HODI. Alternatively, we won't manage it before the release and I then need to correct the GEO restriction to be "one order stricter".