[MITgcm-support] MPI_Recv

[Sergio Jaramillo]

Dear MITgcm-support,

I am trying to use the float package in a parallel run but I have had a 
lot of problems. So far I have tracked the errors down to the subroutine 
exch_recv_get_vec.F. It seems that the model never comes out of a loop 
here after calling the subroutine MPI_Recv. I was looking for MPI_Recv.F 
or something similar, but I don't find it anywhere. I notice that is 
called too in several other subroutines, so It must be somewhere. I 
don't know where else to look. Does it really exists?

Thanks a lot,

Sergio Jaramillo
UBC