Eugene,
As you noticed add_procs only add processes to the list of available
processes without trying to setup any connections to them. As a result
when we return from add_proc it is very unlikely that we will be able
to accurately detect any connection problems.
The connections are established later on, when the first message is
sent to a peer. If this message cannot use a BTL, it will try the next
one that reported a possible connection to the peer until no more BTL
are available for the peer. Then, finally we give up and return a
fatal error.
george.
On Dec 11, 2008, at 21:35 , Eugene Loh wrote:
I'm not exactly sure where the fix to this should be, but I think
I've found a problem.
Consider, for illustration, launching a multi-process job on a
single node. The function
mca_bml_r2_add_procs()
calls
mca_btl_sm_add_procs()
Each process could conceivably return a different value --
OMPI_SUCCESS or otherwise. E.g., if there isn't enough room for all
to allocate all the shared memory they need, early processes might
succeed in their allocations while laggards won't.
The fact that some processes fail doesn't bother the BML. It just
loops over other BTLs and, quite possibly, finds another BTL to make
needed connections.
Is this a problem? It seems to me to be, but I haven't yet figured
out what the BML does next. I'm guessing it ends up with a
hodgepodge of BTLs. E.g., A talks to B via sm, but B talks to A via
tcp. And, I'm still guessing, this produces badness (like hangs).
Comments?
_______________________________________________
devel mailing list
de...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/devel
