Hi slimtimmy
I have been involved in several of the MPI Forum's discussions of how
MPI_Cancel should work and I agree with your interpretation of the
standard. By my reading of the standard, the MPI_Wait must not hang and the
cancel must succeed.
Making an MPI implementation work exactly as the
The MPI_Cancel is only supported for receive requests. It's a no-op
for sends. Therefore your request never get canceled and the wait will
block until the request completion ... which means it will block
forever (as there is no receive corresponding to the request).
george.
On Apr 15, 20
I encountered some problems when using MPI_CANCEL. I call
Request::Cancel followed by Request::Wait to ensure that the request has
been cancelled. However Request::Wait does not return when I send bigger
messages. The following code should reproduce this behaviour:
#include "mpi.h"
#include
usi
hi, am beginner in openmpi
could anyone tell me how do i run programs in open mpi on smp.. as in do I
have to do any changes in hostfile ... like in mpich -- I write
server:2
how do I do it openmpi
Hi Lars,
First off, I think Jeff makes some very good points.
If you still think your applications will benefit from yielding
instead of hogging the cpu,
you should probably try to use the parameter "mpi_show_mca_params".
This will give you a list of the mca parameters at runtime. This way
