Parameter mis-matches in source

[cponder]

I'm building with the gcc/14.2.0 compiler, which is likely stricter than what you've been using for testing.
The GCMv12-rc15 tag gives me 58 of these warnings

[  8%] Building Fortran object src/Shared/@MAPL/gridcomps/History/CMakeFiles/MAPL.history.dir/Sampler/MAPL_StationSamplerMod.F90.o
/work/01768/ponderc/vista/NASA/2025-05-06.GEOSgcm/GEOSgcm/src/Shared/@MAPL/gridcomps/History/Sampler/MAPL_StationSamplerMod.F90:661:24:

  622 |                   p_rt_2d, recvcount, displs, MPI_REAL,&
      |                  2      
......
  661 |                         p_rt_3d_aux(1,k), recvcount, displs, MPI_REAL,&
      |                        1
Warning: Element of assumed-shape or pointer array as actual argument at (1) cannot correspond to actual argument at (2)

and 9 of these

[ 10%] Building Fortran object src/Shared/@GMAO_Shared/GEOS_Shared/CMakeFiles/GEOS_Shared.dir/G3_MPI_Util_Mod.F90.o
/work/01768/ponderc/vista/NASA/2025-05-06.GEOSgcm/GEOSgcm/src/Shared/@GMAO_Shared/GEOS_Shared/G3_MPI_Util_Mod.F90:1055:33:

 1055 |                  call mpi_recv ( buf,imx*jmy*lm,mpi_rkind,npe,npe,comm,status,ierror )
      |                                 1
......
 1595 |           call mpi_recv ( buf,lattice%im(n)*jm,mpi_rkind,peid,peid,lattice%comm,status,ierror )
      |                          2       
Warning: Type mismatch between actual argument at (1) and actual argument at (2) (REAL(8)/REAL(4)).

Likely this isn't the full set, since the build fails at 65% as reported in issue #928. The v11.7.1 build gets further but actually fails due to similar:

[ 76%] Building Fortran object src/Components/@GEOSgcm_GridComp/GEOSogcm_GridComp/@GEOS_OceanGridComp/MOM6_GEOSPlug/mom6_cmake/CMakeFiles/mom6.dir/__/@mom6/src/parameterizations/vertical/MOM_CVMix_KPP.F90.o
/work/01768/ponderc/vista/NASA/2025-05-06.GEOSgcm/GEOSgcm/src/Components/@GEOSgcm_GridComp/GEOSogcm_GridComp/@GEOS_OceanGridComp/MOM6_GEOSPlug/@mom6/src/parameterizations/vertical/MOM_CVMix_KPP.F90:1258:24:

 1258 |                   bfsfc=surfBuoyFlux,                & ! surface buoyancy flux [m2 s-3]
      |                        1
Error: Rank mismatch in argument ‘bfsfc’ at (1) (rank-1 and scalar)

/work/01768/ponderc/vista/NASA/2025-05-06.GEOSgcm/GEOSgcm/src/Components/@GEOSgcm_GridComp/GEOSogcm_GridComp/@GEOS_OceanGridComp/MOM6_GEOSPlug/@mom6/src/parameterizations/vertical/MOM_CVMix_KPP.F90:1296:26:

 1296 |                     bfsfc=surfBuoyFlux, & ! surface buoyancy flux [m2 s-3]
      |                          1
Error: Rank mismatch in argument ‘bfsfc’ at (1) (rank-1 and scalar)
gmake[2]: *** [src/Components/@GEOSgcm_GridComp/GEOSogcm_GridComp/@GEOS_OceanGridComp/MOM6_GEOSPlug/mom6_cmake/CMakeFiles/mom6.dir/build.make:1859: src/Components/@GEOSgcm_GridComp/GEOSogcm_GridComp/@GEOS_OceanGridComp/MOM6_GEOSPlug/mom6_cmake/CMakeFiles/mom6.dir/__/@mom6/src/parameterizations/vertical/MOM_CVMix_KPP.F90.o] Error 1
gmake[1]: *** [CMakeFiles/Makefile2:24845: src/Components/@GEOSgcm_GridComp/GEOSogcm_GridComp/@GEOS_OceanGridComp/MOM6_GEOSPlug/mom6_cmake/CMakeFiles/mom6.dir/all] Error 2
gmake: *** [Makefile:166: all] Error 2

[mathomp4]

@cponder I think this bug might be due to an old mepo. Can you check if you have mepo 2.3.2 via:

mepo --version

[cponder]

If I use mepo 2.3.2 I can get a full build of the v11.7.1 but the GCMv12-rc15 still fails at the same place.
That said, there are 51 warnings of the first type and 15 of the second type in the source, just no fatal failure with v11.7.1.
I do worry that there could be be pointer failures or corrupted data at runtime.