search

phil-blain / CICE

Development repository for the CICE sea-ice model
Other
0 stars 0 forks source link

performance: check if allocating stress_Pr as (nx_block,ny_block,max_blocks,8) instead of (nx_block,ny_block,8) is faster #41

Open phil-blain opened 2 years ago

phil-blain commented 2 years ago

same for diag_rheo. They are declared private in the OpenMP loops which means that the whole arrays is allocated when entering the loop?