[mvapich-discuss] one-sided passive communications

["María J. Martín"]

Hi Jim,

> The async progress threads (one is spawned for every MPI process) are implemented as a Wait() on an Irecv() control message that will be sent when MPI_Finalize is called.  Depending on the underlying implementation (maybe an MVAPICH expert could comment here), this could poll and consume a core.
> 
> Is it possible that the application is running very slowly, rather than deadlocking?  


Yes, It is possible. This fits with the fact that the application runs successfully when extra cores are allocated.

> Another debugging suggestion would be to attach a debugger or generate a core file during the hang to give some more insight into where the processes are stuck.

We will try. 

Thank you very much for your suggestions.

Best,

María