marcadella

Hi, Any progress on this task?

Note that `do_reset_veg` et al were removed from master in v5.0 as they were not used. They can be re-added as needed when incorporating disturbance support (i.e. when merging the...

A better test for finding arrays with lower bound is the regexp `(?=.*::)(?=\(\s*\d+\s*:)|\(\s*0\s*:` It will detect both use of 0 index and the definition of arrays with lower bounds.

All lower bounds have been exterminated (but one which should not be a problem as it is used in a function.)

Ideas of what the problem could be: 1. Shared library mismatch: we compile with a set of libraries and run on a platform with a different set of libraries (different...

Update: - Compiling and running `analysis/biome_segmentation_fault.R` works fine (no segfault) on Ubuntu 24.04 with R 4.4.1 and gcc 13.2.0. - Compiling and running `analysis/biome_segmentation_fault.R` gets segfault on Ubuntu 24204 with...

To sum up: - I think the Segfault came from a buggy gcc version and updating both gcc and gfortran to v12 or more should fix those. I observed it...

With gcc v12 and v13, I get completely random simulation outputs... I am 99% sure that there is actually an issue in Fortran (like bad pointer usage or something like...

I found one issue and it might actually explain it all. In `initialize_vegn_tile`, `pleaf`, `proot` and `pseed` are never initialized. They take therefore random values propagating through the simulation. It...

There some more uninitialized variables (for instance `pseed%n%n12`, `pmicr` both c12 and n14). I think the best strategy is if @lauramarques could go through the `initialize_vegn_tile` subroutine and check the...