[MITgcm-support] monthly forcing data

[Martin Losch]

Hi Estanislao,

That’s consistent with my experience with dynamic allocation in fortran in general (not with the MITgcm in particular): different compilers work differently and dynamic array allocation is usually the first thing that goes wrong (probably because as a programmer you can make more mistakes there that are sometimes tolerated and sometimes not). So it makes sense that something that works on one cluster with one compiler, doesn’t work with a different compiler.

Martin

> On 5. Apr 2022, at 16:33, ESTANISLAO GAVILAN PASCUAL-AHU <20190929 at hhu.edu.cn> wrote:
> 
> 
> Hi Martin,
> 
> you were right.  As you suggested I turned off the EXF_INTERP_USE_DYNALLOC, and then I changed the exf_max_nlon/nlat and exf_interp_bufferSize in  EXF_INTERP_SIZE.h. Now the model works perfectly fine. This issue is still curious because I did not have any problem in a different cluster. In fact, the only differences between the clusters were the version of the intel compilers (2017u5 vs 2018.1) and the netcdf libraries. Thanks again for all your help.
> 
> Kind regards,
> 
> Estanislao
> 
> 
> 
> 
> 
> 
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.mitgcm.org/pipermail/mitgcm-support/attachments/20220407/489a9d2c/attachment.html>