When running tests for PR #477, I noticed that I could not run tests with one particular suite (SCM_GFS_v16_RRTMGP/"mallard"). Apparently it wasn't my changes though, because the same problem appears with the top of main: All the test cases I tried (astex, twpice, bomex) ended with the same failure:
SURFACE EMISSIVITY INDEX, IDM,JDM: 360 180 NOTE: DATA FROM N TO S
Reading RRTMGP shortwave k-distribution metadata ... "
stderr: " RRTMGP implicit cloud scheme being used.
Calling ccpp_physics_init with suite 'SCM_GFS_v16_RRTMGP'
In file '/Volumes/d1/kavulich/CCPP/rename_suite_files/after_rebase/main/ccpp-scm/ccpp/physics/physics/Radiation/RRTMGP/rrtmgp_sw_gas_optics.F90', around line 226: Error allocating 12519926383860640000 bytes: Cannot allocate memory
Unfortunately my Mac laptop does not have 12.5 exabytes of memory available, so the run segfaults.
Code was built on MacOS Ventura 13.6.7 with AppleClang 15.0.0.15000100 (C compiler) and GNU 12.3.0 (fortran compiler), with openmpi-4.1.6 from spack-stack 1.6.0.
When running tests for PR #477, I noticed that I could not run tests with one particular suite (SCM_GFS_v16_RRTMGP/"mallard"). Apparently it wasn't my changes though, because the same problem appears with the top of main: All the test cases I tried (astex, twpice, bomex) ended with the same failure:
Unfortunately my Mac laptop does not have 12.5 exabytes of memory available, so the run segfaults.
Code was built on MacOS Ventura 13.6.7 with AppleClang 15.0.0.15000100 (C compiler) and GNU 12.3.0 (fortran compiler), with openmpi-4.1.6 from spack-stack 1.6.0.