Ocean Modeling Discussion

Never mind! I have figured this out! I needed to add "-lnetcdf" to the linker command! Not sure why that didn't happen but I guess it doesn't matter at this point.

Hi ROMS people, We've recently been trying to compile ROMS with PIO on the cluster we are using. I had little problem compiling pnetcdf and PIO (using the instructions here https://www.myroms.org/wiki/External_Libraries) as a guidline. However, when I attempted to compile ROMS (most recent svn ...

I am having trouble starting floats in the middle of a run. I want to re-start a run (that was run without floats) and then add floats at a few key locations and view their evolution. So, I recompiled with #define FLOATS and the initial header info shows that my floats.in file is being read ...

Okay I figured out how to resolve this. If I copy mpiif.h directly to the Build directory manually it will compile. I still can't figure out why it doesn't see it in my mpich2/include directory though, especially in light of the fact that it works fine under PGI.

thanks, David

jcwarner wrote:check your environment and see if the location of mpif.h is in your INCLUDE list.

That's the bizarre thing. It is. I should have mentioned this. I had no problem, by the way, of getting the PGI version of this to compile with MPICH2.

David

I can't seem to get ROMS3 to compile with gfortran and MPICH2. I have no trouble compiling the serial version. I have compiled (and recompiled) the MPICH2 package with gfortran (with no apparent problems) but when I attempt to compile ROMS I consistently get: cd Build; gfortran -c -frepack-arrays ...