1
1
openmpi/orte/mca/routed/tree
Ralph Castain ec5fe78876 When in the unity message routing mode, we have to update the RML contact info in the parent procs so that they know how to talk to the children. Ideally, this would be done in the MPI layer since that layer knows which procs are actively involved in the comm_spawn. However, it isn't being done there, which causes comm_spawn to fail, so do it explicitly in the RTE.
Note that this means ALL procs in the parent job are updated, even though they may not be participating in the comm_spawn. This doesn't really hurt anything - just unnecessary.

Comm_spawn still has a problem when a child process shares a node with a parent, so this doesn't fix everything. It only fixes the bug of ensuring all procs know how to talk to each other.

This commit was SVN r16460.
2007-10-16 16:09:41 +00:00
..
configure.params A number of improvements / changes to the RML/OOB layers: 2007-07-20 01:34:02 +00:00
Makefile.am A number of improvements / changes to the RML/OOB layers: 2007-07-20 01:34:02 +00:00
routed_tree_component.c When in the unity message routing mode, we have to update the RML contact info in the parent procs so that they know how to talk to the children. Ideally, this would be done in the MPI layer since that layer knows which procs are actively involved in the comm_spawn. However, it isn't being done there, which causes comm_spawn to fail, so do it explicitly in the RTE. 2007-10-16 16:09:41 +00:00
routed_tree.c Squeeeeeeze the launch message. This is the message sent to the daemons that provides all the data required for launching their local procs. In reorganizing the ODLS framework, I discovered that we were sending a significant amount of unnecessary and repeated data. This commit resolves this by: 2007-10-11 15:57:26 +00:00
routed_tree.h When in the unity message routing mode, we have to update the RML contact info in the parent procs so that they know how to talk to the children. Ideally, this would be done in the MPI layer since that layer knows which procs are actively involved in the comm_spawn. However, it isn't being done there, which causes comm_spawn to fail, so do it explicitly in the RTE. 2007-10-16 16:09:41 +00:00