Closed vanroekel closed 7 years ago
@mark-petersen this bug came up with the new 18to6 80 layer configuration where we made the minimum depth smaller. It is now at the 10m threshold for the mixed layer depth calculation, resulting in an unitialized variable. here is output from a test on grizzly with intel, debug mode
forrtl: severe (194): Run-Time Check Failure. The variable 'ocn_mixed_layer_depths_mp_ocn_compute_mixed_layer_depths_$REFINDEX' is being used in 'mpas_ocn_mixed_layer_depths.F(258,16)' without being defined
Image PC Routine Line Source
ocean_model 0000000002D35844 ocn_mixed_layer_d 258 mpas_ocn_mixed_layer_depths.F
ocean_model 00000000026F74B7 ocn_analysis_driv 1192 mpas_ocn_analysis_driver.F
ocean_model 00000000026EC14B ocn_analysis_driv 593 mpas_ocn_analysis_driver.F
ocean_model 00000000026E2A8A ocn_forward_mode_ 481 mpas_ocn_forward_mode.F
ocean_model 000000000276FD87 ocn_core_mp_ocn_c 111 mpas_ocn_core.F
ocean_model 0000000000414460 mpas_subdriver_mp 338 mpas_subdriver.F
Stack trace terminated abnormally.
--------------------------------------------------------------------------
mpirun detected that one or more processes exited with non-zero status, thus causing
the job to be terminated. The first process to do so was:
Process name: [[50981,1],394]
Exit code: 194
--------------------------------------------------------------------------
With this PR, the problem is fixed and the error does not occur.
Looks good. I compiled to be safe.
For RRS18to6 there are a few columns where refIndex can be uninitialized in the mixed layer depths AM. This fixes that error.